+ All Categories
Home > Documents > Remittances in the Pacific An Overview

Remittances in the Pacific An Overview

Date post: 01-Jan-2017
Category:
Upload: lyngoc
View: 213 times
Download: 0 times
Share this document with a friend
80
Remittances in the Pacific An Overview John Connell Richard P.C. Brown March 2005
Transcript

Defining Hardship and Poverty i

Remittances in the PacificAn Overview

John ConnellRichard P.C. Brown

March 2005

Remittances prelim.pmd 15/03/2005, 11:01 AM1

Pacific Studies Series

The series is published by the Asian Development Bank to provide thegovernments of its Pacific developing member countries with analyses ofeconomic and other issues. The studies are also expected to shed light on theproblems facing governments and the people in the Pacific islands and to suggestdevelopment strategies that combine both political and economic feasibility. Acomplete list of titles in the series is available on ADB’s website (www.adb.org).

© 2005 Asian Development Bank

All rights reserved. Published 2005.Printed in the Philippines.

Library of Congress Cataloging-in-Publication Data Available.

Publication Stock No: 110904

The views expressed in this paper are those of the author and do not necessarilyreflect the view or policies of the Asian Development Bank or its Board ofGovernors or the governments they represent.

The Asian Development Bank does not guarantee the accuracy of the dataincluded in this publication and accepts no responsibility for any consequenceof their use.

Use of the term “country” does not imply any judgment by the authors or theAsian Development Bank as to the legal or other status of any territorial entity.

Remittances prelim.pmd 15/03/2005, 11:01 AM2

able of ContentsTForeword vExecutive Summary vii

1. Introduction 1

2. The Importance of Remittances 5

3. The Composition of Remittances 14

4. The Rationale for Remittances 17

5. The Duration of Remittances 23

6. The Use and Impact of Remittances 30

Debt Repayment 30Consumption 31Savings 32Air Fares and Education 33Investment 33Community Use 35Social Uses 36

7. The Disincentives of Remittances? 38

8. The Policy Context 42

Policies to Foster Remitting through Official Channels 45Policies to Foster Savings and Investment 47Monetary and Anti-inflation Policies 51

9. Conclusions 53

References 61

Remittances prelim.pmd 15/03/2005, 11:01 AM3

Remittances prelim.pmd 15/03/2005, 11:01 AM4

The research for this report was initiated by Andrea Iffland, who renderedtechnical support and guidance during the ensuing phases of in-depth reviews.John Connell, School of Geosciences, University of Sydney and Richard P.C. Brown, School of Economics, University of Queensland prepared thereport based on an extensive literature review. Judy Goldman providededitorial support and much appreciated advice.

The report provides an overview of labor migration from Pacific islandcountries and the remittances that follow. Remittances have always beenand continue to be a major feature of the Pacific island economies. Theyensure a balance of payment surplus in most receiving countries. Thedependence on remittances, particularly of Polynesian countries, iscontroversial. Some economists consider reliance on these private flows offunds critical as it leaves economies vulnerable to fluctuations outside thecontrol of governments. There is also very little understanding about thereciprocity triggered by remittances and its impact on the balance of paymentsand on the gross national product. The lack of reliable and accurate data onremittances renders it difficult to address these concerns. The situation isfurther aggravated as remittances are underestimated due to informalunrecorded flows. Governments recognize the need for better data but arealso aware of the sensitivity of these flows to official monitoring.

Remittances constitute a major part of household cash income for manyfamilies. They are considered a safety net for the population at large.Remittances are mainly used for consumption while investment purposesare of lesser importance. With private sector development a priority of mostPacific island countries, it is essential to understand the remittance flows—their vulnerability, sources, and volume—and how best to harness them fordevelopment purposes.

orewordF

Remittances prelim.pmd 15/03/2005, 11:01 AM5

Remittances prelim.pmd 15/03/2005, 11:01 AM6

Migration is very significant in Pacific island states, especially inPolynesia, primarily as a response to uneven economic and socialdevelopment. In many Pacific island countries, the remittances that flowfrom internal and international migrants to family members at home havebeen of growing importance, again especially in Polynesia where they oftenrepresent the single most prominent component of national incomes. Theyreach levels rarely found elsewhere in the world.

Very few detailed studies of this migration-remittance nexus have beenstatistically significant, and many are more than a decade old so may nolonger be valid. Information in many areas is inadequate, hence furtherresearch would be valuable. Much of the available data concern Samoa andTonga, where international migration and remittances have been of greatimportance and where this has been documented more substantially andcredibly than elsewhere. To some extent, these two Polynesian states providea template of a remittance-dependent economy that can be used as a goodstarting point for other small countries in the region. As long as seriouseconomic challenges face island states, as population growth rates remainabove world averages, and as expectations rise, the ability to migrate will becrucial where development prospects are few and where the possibilities ofdeclining aid levels are becoming more apparent. In fact, many islandgovernments actually promote international labor migration.

Remittances are bi-directional because social networks continue to besignificant. As long as migrants and their kin build and maintain thesenetworks, remittances are likely to be sustained beyond what economicprinciples might suggest. They respond to an implicit social contract,contribute to human capital formation, and can be seen as a form ofintergenerational transfer. Remittances in the Pacific currently continue athigh levels for very long periods except when close kin die or when familiesreunite in the host country. Second generation migrants, however, are likelyto send smaller amounts only on demand. Maintaining remittance flows athigh levels therefore requires a steady flow of new migrants.

There is no question that remittances have contributed to developmentin various contexts. They meet real needs especially on smaller islands, inremote regions, and in independent states. Remittances have tended to go

xecutive SummaryE

Remittances prelim.pmd 15/03/2005, 11:01 AM7

Assessment of Hardship and Poverty in PDMCsviii

into consumption (including house construction) but are increasingly directedtowards investment (especially in the service sector) despite limitedopportunities. Remittances are used to create human capital for futuremigration and do not crowd out or disadvantage sectors of national economies.If this investment in human capital is treated as a legitimate and rational useof remittances, many apparent problems disappear and migration becomesmore obviously beneficial. Cross-sectional data from recent studies showfew signs of remittance decay with length of absence from the home countryand that an increasingly significant motivating factor for migrants to remit isthe accumulation of assets and investments at home. Both of these conclusionsare contrary to many previous studies of migration and remittances in thePacific and elsewhere.

International migration has had both positive and negative effects in thePacific, but the positive effects of notably increased standards of living mustbe contrasted with the limited development potential of many countries andtheir failure to achieve significant, sustainable economic growth. Maximizingthe benefits from international migration is crucial, but Pacific island stateshave not developed policies to do so though opportunities exist to stimulateand direct flows. The present conclusions suggest that remittance levels wouldbe sensitive to policies affecting relative real interest rates but do not suggestthat policies would stimulate more domestic investment or that the migrantsor their families would necessarily make the best entrepreneurs. Policiesshould be geared more towards encouraging migrants to become more activein domestic capital markets as saver-rentiers. This raises the broader issue ofthe general investment climate. Migrants are unlikely to risk their capital inan investment in the home economy if safer alternatives exist elsewhere.Governments must offer savers competitive real interest rates in order toaccumulate loan funds to invest either in domestic projects or to hold asoverseas assets.

Remittances prelim.pmd 15/03/2005, 11:01 AM8

Introduction

1

In many Pacific island countries, the remittances that flow from internaland international migrants to family members at home are a prominent featureof the national economy. Internal rural-urban migration produces some flowsof goods and money, but international migration yields substantially greaterones. There have, however, been few studies of this migration-remittancenexus and even fewer are both detailed and statistically significant. Manyare now more than a decade old and may no longer be valid. Information istherefore somewhat limited on the current use of remittances, on their realand potential contribution to domestic incomes and economic development,on their social influence, and on how they function as a safety net againsthardship and poverty. Furthermore, many additional questions have not beenthoroughly investigated including the extent to which remittances are sensitiveto variables like foreign exchange rates, relative rates of interest and inflation,and the extent to which migration and remittances contribute to inequality.Remittances have raised living standards in the islands, they have contributedto employment especially in the service and construction sectors, and theyhave eased balance of payment problems despite contributing to inflationespecially in the larger Polynesian countries. However, it has been arguedthat remittances have not generally been invested in economic growth, soincreased demand for improved consumer goods can usually be met only byfurther migration. This particular issue is the principal theme of this overview.

In the Pacific island microstates, the prospects for economic growth areunusually limited compared with other regions of the world. Consequently,the now widely perceived disparities in economic development and welfarebetween them, especially the smallest states of Polynesia and Micronesia,and neighboring developed nations have contributed not only to substantialmigration but also to increasing pressures for further migration. Almost allinternational migration has occurred since the 1960s, but it has become soextensive that some of the greatest concentrations of Pacific islanders are incities such as Auckland, Honolulu, and Los Angeles rather than in the islandPacific. Already there are second and third generations of islanders for whom“home” is sometimes an uncertain and ambivalent concept.

Introduction1

Pacific-Remittances.pmd 15/03/2005, 10:56 AM1

Remittances in the Pacific

2

As migration continues, small, vulnerable Pacific states have become aperipheral and dependent part of a wider world. Contemporary patterns ofmigration have diversified, demographic structures have changed, and globaland island economies present different development contexts. Essentially,the lives of island people are increasingly international. Some small Polynesianstates like Cook Islands and Niue have more than half their populations livingoverseas. The largest migratory streams have been from Polynesia thoughthere is now increasingly rapid migration from the independent Micronesianstates that were once American colonies. Migration has been particularlysubstantial and better documented from Samoa (formerly Western Samoa)and Tonga—the two largest independent Polynesian states. In contrast, forAmerican Samoa, Palau, Tokelau, Wallis and Futuna, and others, there isvery little contemporary information despite the significance of migrationand remittances.

Migration in the Pacific is primarily a response to real and perceivedinequalities in socioeconomic opportunities that have resulted from unevendevelopment. Social influences on migration are important especially in termsof access to education and health services but also as one element of thetransition to adulthood; such social influences are in turn often a function ofeconomic issues.

Basically, migration is a time-honored strategy of moving from a poorarea to a richer one in search of social and economic mobility. In most ofindependent Polynesia and Micronesia there is a powerful and almost universalrecognition that the best social and economic opportunities lie overseas. Infact, the desire to migrate overseas is now regarded as normal though mostmoves began only half a century ago. Thus in Tonga, quite simply, “…overallthere are few opportunities for socioeconomic advancement in Tonga andmigration is perceived as the only solution” (Morton Lee 2004). Much thesame might be said of smaller states. What is also true everywhere is thatwhile rural-urban migration produces remittances, international migrationproduces substantially greater flows and is therefore more attractive.

Radical changes in perceptions of satisfactory standards of living,desirable occupations, and adequate services and amenities have encouragedmigration. Aspirations have almost always included imported food and othergoods (such as clothes and vehicles) and access to schools, hospitals, andmodern entertainment, all of which demand some cash income (Bedford1980). At the same time, agricultural work throughout the Pacific has beenlosing prestige, and young men have been less willing to participate. Changingaspirations have not only been the province of young men. In Tonga, “… oneoften hears parents expressing the wish that their children work at somethingbetter than agriculture even though they themselves are farmers. This

Pacific-Remittances.pmd 15/03/2005, 10:56 AM2

Introduction

3

something better invariably refers to white collar jobs which carry with thema lot of prestige” (Sevele 1973). In the past 25 years, these trends have becomewell established throughout the region.

Changes in values following increased educational opportunities and thegrowth of bureaucratic (largely urban) employment in the region in the 1970s,especially during the prelude to independence in many states, furtherencouraged migration as local employment opportunities did not keep pacewith population growth. This situation continued through the 1990s, anothera period of both rising formal unemployment and restructuring in much ofthe region. This widened the gap in many places between rising expectationsfor consumer goods and travel and the reality of limited domestic employmentand incomes.

In terms of international migration, what has distinguished most Pacificislanders is that they intend to stay in the host country even though they mayhave expressed an intention to return home (Macpherson 1994). This isparticularly true of most migrants from Samoa and Tonga. It is also true ofmigrants from Cook Islands and Niue despite the fact that they are citizensof New Zealand and can move freely between countries. Because of thistendency, remittances from Pacific islanders are particularly substantial. Otherthan for Kiribati and Tuvalu, direct comparisons with other states with amigration-remittance nexus such as Pakistan or the Philippines are limitedbecause most of their migrants are on short-term contracts and must returnhome. In contrast, until quite recently when significant migration from Tuvaluto New Zealand began, almost all migration from Kiribati and Tuvalu waseither as laborers to nearby Nauru or as contract workers on shipping lines.Workers were specifically trained for these activities, their remittances wereand are both obligatory and institutionalized, and they were required to returnhome.

Remittances play an important role especially when migration is fromsmall island states such as Tuvalu or from small islands (such as Ponam orWare) in larger Pacific states like Papua New Guinea. Otherwise remittanceswithin countries, especially in Melanesia, are of limited economic significancethough they may be of considerable social significance (Mecartney 2000).However, even for remote parts of such large islands as Tanna, Vanuatu,remittances from urban migrants are the single largest source of income inseveral villages (Winthorpe 2004). Nonetheless, in many places there isinadequate information on remittances and their use and hence on their realand potential contribution to domestic incomes and economic development,their influence on social change or continuity, and their ability to constitute asafety net.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM3

Remittances in the Pacific

4

Because of the continued significance of remittances, the sustainabilityof remittance-dependent development is particularly important and somewhatuncertain especially if in the home countries the need for remittances growsfaster than the supply (Forsyth 1992; Macpherson 1992). The rate of migrationto major destinations, namely New Zealand, Australia, and the United States,has declined in recent years due to economic recessions in those countriesand to the restructuring of migration regulations such that return migrationhas sometimes been considerable (Maron 2001). At the same time, migrationpolicies in host countries have increasingly favored those with skills (Liki2001; Brown and Connell 2004). Furthermore, as families reunite overseasand as migrants integrate into host communities, their ability and willingnessto remit are expected to decline over time (e.g. Macpherson 1994). If that isso, unless other sources of income can be developed and unless remittanceincomes can be used and invested more productively, the future of theeconomies of remittance-dependent Pacific countries is uncertain.

Several states in the region actually have an outflow of remittances. Morethan half the population of the Commonwealth of the Northern Marianas isnonresident workers. Their remittances, mainly to the Philippines and to alesser extent to China, totalled about US$80 million in 2003. Workers onNauru also remit to their home countries including Kiribati and Tuvalu thoughthere are no data on these or other flows, and there is a significant flow ofremittances from New Caledonia to Wallis and Futuna. Remarkably there isalso a flow from Niue because of extremely high levels of aid, the assistancethat Niueans provide to those resettling in New Zealand, and perceptionsthat the community will not remain so it is useless to remit to Niue (Cohn2003; Heyn 2003). These rather exceptional cases are not discussed herethough that of Niue has important implications.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM4

Introduction

5

There is considerable evidence in many countries, at least in past years,that migrants’ remittances are a significant part of disposable incomeespecially in Samoa, Tonga, and smaller states such as Cook Islands, Kiribati,Tuvalu, and Wallis and Futuna. In fact, the smaller island countries(specifically Cook Islands, Kiribati, Tokelau, and Tuvalu) were dubbedMIRAB states, where migration, remittances, aid and the resulting, largelyurban bureaucracy are central to the socioeconomic system (Bertram andWatters 1985). It quickly became evident that this acronym had widerrelevance (Connell 1988). While the term is disliked in the Pacific because itimplies a handout mentality, it nonetheless suggests the centrality of migrationand remittances and has been largely unchallenged for two decades (Bertram1999a). On the other hand, it has been recognized quite correctly that theterm appears to deny the initiative of Pacific islanders other than as migrantsin spite of a range of activities such as the 1990s boom in squash cultivationin Tonga (van der Grijp 1999; cf. Hooper 1993). Moreover,

Islanders in their homelands are not the parasites on theirrelatives abroad that misinterpreters of ‘remittances’ would haveus believe. Economist do not take account of the social centralityof the ancient practice of reciprocity. ... They overlook the fact thatfor everything homeland relatives receive they reciprocate withgoods they themselves produce, by maintaining ancestral roots andlands for everyone. ... This is not dependence but interdependence(Hau’ofa 1984)

At the very least, remittances have complex and important social andeconomic dimensions.

The original acronym has spawned others. MURAB places extra andappropriate emphasis on urbanization in island states such as Tuvalu (Munro1990), and ARAB applies to French Polynesia where there is virtually nomigration because of high wage levels and other forms of state support butwhere atomic rent produced similar outcomes (Poirine 1994). Finally therehave also been MIAB countries where migration has not stimulated asignificant flow of remittances (Ogden 1994).

The Importance of Remittances2

Pacific-Remittances.pmd 15/03/2005, 10:56 AM5

Remittances in the Pacific

6

MIAB describes some parts of Micronesia including the Marshall Islandsand the Yapese (Federated States of Micronesia) outer island of Woleai (Ogden1994; Karakita 1997). Indeed more remittances flow from the Marshall Islandsto migrants in the United States than into the islands. This is a function ofrelatively high incomes because of American aid, high education expensesin the United States (Hess 2004), and the fact that Marshallese haveexceptionally low-paying jobs in the United States. This situation is also truefor some parts of the Federated States of Micronesia (FSM). Migrants fromthe atoll of Namoluk do not send remittances home on a regular basis sincethe cost of living in California, Hawaii, and Guam is such that they havenothing to send. Moreover, with few exceptions they work in notoriouslypoorly paid, largely forgotten jobs (Marshall 2004). However, as Micronesiansacquire superior education and better jobs, it is probable that the migration-remittance system will eventually move towards the more standard form. Infact, the largest survey of remittances of migrants from FSM to Guam andHawaii in the late 1990s demonstrated that more than a third of householdsdid send them (Grieco 2003). The proportions were lower than those of firstgeneration Polynesian migrants and the sums were also smaller, but it isclear that remittances are of growing significance. Connell’s observations onthe atoll of Woleai in the early 1990s suggest that this transition was alsooccurring there as other sources of cash income were both few and declining.

Remittances are even more important in the more remote parts of smallPacific island states whether as international or national flows. On the tinycoral atoll of Manihiki in Cook Islands, migration and remittances constitutednothing less than a socioeconomic strategy for collective survival (Underhill1989). In Nanumea in Tuvalu, remittances grew from half the island’s incomein the 1970s and 1980s to 75% in the 1990s in large part because of thecollapse of copra marketing as world prices slumped (Chambers and Chambers2001). Similar situations occur in other small islands and island states.

In Tonga and elsewhere, migrants have been seen as part of a“transnational corporation of kin” that may seek to maximize extendedhousehold incomes across different continents (Marcus 1981). Similarly forhouseholds in Samoa, “…having young wage earners abroad diversifiedfamilies’ earning streams and reduced their dependence on high-risk activities.Having family members in several locations abroad diversified earning sourcesand reduced risk levels still further” (Macpherson 2004). MoreoverMacpherson went on to argue that, “Families, using intelligence from migrantsabroad, periodically surveyed risks and returns in various enclaves andencouraged others abroad to relocate in places in which returns were foundto be higher and risks lower.” In this way Samoans were encouraged to jointhe United States military because jobs were assured, wages were higher,

Pacific-Remittances.pmd 15/03/2005, 10:56 AM6

The Importance of Remittances

7

and education could be obtained without loss of earnings. “If this analysisdepicts Samoans as calculative and instrumental, it is because in relation torisk and return they are necessarily so... [as] risks and returns available invarious places were formally canvassed and modeled by families” (ibid).

It has been argued, however, that applying the same kind of model inTonga tends to portray families as in agreement about their economic aimsand functions whereas there are often conflicts and tensions within them(Morton Lee 2004). Moreover, over a decade ago, James argued that in manyTongan villages remittances were becoming individualized and that the ideaof a transnational community of kin was becoming increasingly invalid (James1993b; Morton Lee 2003). In Kiribati and Tuvalu where seafarers sendremittances to wives and parents, there are frequent disagreements about theirallocation and use (Dennis 2003). Evans has pointed out that since remittancesnecessarily tend to be received by a single person, their use remains morecomplex than just meeting certain household goals (2001). In the absence ofdetailed studies in the Pacific in the last decade, the extent to which there hasbeen greater individualization is impossible to determine. Such conflicts overuse emphasize, rather than downplay, the role of remittances.

Only recently have second and third generations of Pacific islanders grownup outside their island homes, hence the extent to which these people willremit and even whether they can be described as islanders or migrants is notwell known. Already there is some evidence that the links between secondgeneration Samoans in New Zealand and Tongans in Australia with their kinon home islands are declining though in the latter case links are maintainedwith migrant Tongans elsewhere (Muliaina 2001; Morton Lee 2003; 2004a).These new generations are more likely to act as individuals rather than toperceive themselves as members of wider transnational social groupings.

Until the 1990s, there was a degree of consensus about migration andremittances in the Pacific centered on their high volume, their socialsignificance, and their relatively conservative use for consumption and socialevents. While it is generally accepted that remittances are an important benefitof migration, at the same time they have been perceived as both central toand an element of distortion and dependence in island economies. Chargeshave been laid that remittances fuel conspicuous consumption of importedgoods of little real value, that they fail to transform or rejuvenate languishingrural economies, and that they discourage local effort. Pacific islanders, onthe other hand, have often viewed remittances more positively. In 1992Malama Meleisea wrote,

In less than twenty years Samoan villages have been rebuiltwith permanent material houses, shiny glass and concrete churches.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM7

Remittances in the Pacific

8

Many rural families own smart new Japanese or American madepick-up trucks and young people wear jeans and shorts instead oflavalava – all this has been due mainly to the remittance dollar.

Similarly in Tonga,

The consequences of having a relative overseas were oftenconspicuous in daily life in the early 1980s. New store boughtoutfits. A household appliance, or frequent trips to the village shopwere local signs that a letter or package had come from overseas.A house in the village with a sink or bath-tub, sometimes sittinguninstalled on the property, was an indication that a husbandworking overseas would soon be returning (Small 1997).

Though incomes from outside the islands (whether remittances or aid)have tended to contribute to inflation, to raise the opportunity cost of labor,and possibly to depress some local development, they have nonethelessproduced higher material standards of living where the local resource bases,fragmentation, and isolation would otherwise have impeded development.In many circumstances, remittances are very high proportions of national,village, and household incomes. For the Samoan economy, for example,remittances are a crucial element in national development plans. Even asearly as 1962, Pirie and Barrett predicted that migration and remittanceswould become central to Samoan economic planning. Forty years ago, Pirieobserved that Samoans in New Zealand had remitted US$15 million and thatthis was, “...of major assistance to Western Samoa in meeting its overseaspayments” (1960). At much the same time, Shankman recorded that, “…in1974, 20 percent of the Western Samoan population was overseas andremitting over 50 percent of the national income” (1976). This situation wasbroadly similar in Tonga where remittances were the single most importantsource of national income at least by 1973 (Connell 1983) although theseearly trends were less well documented (see Campbell 1992). In Cook Islands,high proportions were also recorded though aid played a greater role (Loomis1990; Mataio 1991). In each of these states, for more than a quarter of acentury since, remittances have remained at similarly high proportions orhave increased.

The overall data on remittances as gross private transfer receipts (seefollowing table) though under-recorded indicate the very considerable long-term significance of remittances in Samoa and Tonga and their recent growingimportance in Fiji Islands.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM8

The Importance of Remittances

9

Gro

ss P

riva

te T

rans

fer

Rec

eipt

s a i

n U

S$ m

illio

n 19

90–2

002

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Coo

k Is

land

s b

--

--

1.2

1.3

1.4

1.4

1.2

1.1

1.1

Fiji

Isla

nds

c-

--1

6.3

-11.

1-2

3.8

-26.

5-3

0.4

-11.

4-7

.0-1

0.7

-12.

340

.453

.0K

irib

ati d

--

--

4.0

5.2

6.0

6.8

6.7

7.1

6.0

5.7

6.0

Mar

shal

l Isl

ands

--

0.5

0.5

0.5

0.5

0.5

0.5

0.6

0.6

0.6

0.6

0.6

Mic

rone

sia,

Fede

rate

d St

ates

of

--

--

1.1

1.2

1.4

1.7

1.9

2.2

2.2

2.2

2.3

Papu

a N

ew G

uine

a-

--

-59

.466

.551

.770

.383

.061

.813

.026

.018

.0Sa

moa

e-

--

31.3

33.1

35.3

36.8

42.0

37.3

41.6

43.7

41.8

57.9

Solo

mon

Isl

ands

--

--

-22

.118

.617

.511

.924

.916

.523

.915

.9T

onga

f-

35.5

32.5

41.7

42.5

46.9

44.9

44.4

44.2

40.2

48.0

54.2

65.2

Tuv

alu

g-

1.7

1.7

1.7

2.0

2.0

2.0

4.0

4.0

4.0

4.0

4.9

Van

uatu

6.9

6.0

5.4

4.9

5.8

6.1

12.5

13.8

15.6

18.7

27.3

39.5

30.6

aPr

ivat

e tr

ansf

er r

ecei

pts

in th

e B

alan

ce o

f Pa

ymen

ts a

ccou

nt a

re c

urre

nt tr

ansf

er p

aym

ents

rec

eive

d by

pri

vate

per

sons

and

non

offi

cial

org

aniz

atio

ns o

f th

e re

port

ing

coun

try

from

the

rest

of

the

wor

ld th

at c

arry

no

prov

isio

ns f

or r

epay

men

t. In

add

ition

to w

orke

rs ‘

rem

ittan

ces,

pri

vate

tran

sfer

rec

eipt

s w

ould

incl

ude

curr

ent t

rans

fers

by

mig

rant

s;gi

fts,

dow

ries

, and

inhe

rita

nces

; and

alim

ony

and

othe

r su

ppor

t rem

ittan

ces.

bPr

ivat

e re

mitt

ance

rec

eipt

s on

lyc

Net

Pri

vate

sec

tor

tran

sfer

sd

Ref

ers

to s

eam

en’s

rem

ittan

ces

whi

ch a

re c

lass

ifie

d by

Kir

ibat

i aut

hori

ties

in th

e B

alan

ce o

f Pa

ymen

ts a

s fa

ctor

inco

me

rece

ipts

rat

her

than

tran

sfer

rec

eipt

s.e

Exc

lude

s ea

rnin

gs r

emitt

ed b

y re

side

nts

wor

king

on

shor

t-te

rm c

ontr

acts

ove

rsea

s (t

his

wou

ld b

e de

fine

d as

fac

tor

inco

me

from

abr

oad)

.f

In f

isca

l yea

r en

ding

30

June

.g

Est

imat

es fo

r the

ear

ly 1

990s

, 199

7 to

200

0, a

nd 2

001

from

the

Cen

tral

Sta

tistic

s D

ivis

ion

and

the

TT

F A

dvis

ory

Com

mitt

ee 2

001

repo

rt a

s ci

ted

in th

e T

uval

u 20

02 E

cono

mic

and

Publ

ic S

ecto

r R

evie

w, A

DB

Man

ila, p

age

9.

Sour

ces:

Tre

asur

y of

Sam

oa Q

uart

erly

Rev

iew

, var

ious

issu

es; N

atio

nal R

eser

ve B

ank

of T

onga

Qua

rter

ly B

ulle

tin; C

ook

Isla

nds

Qua

rter

ly S

tatis

tical

Bul

letin

; Res

erve

Ban

k of

Fiji

Qua

rter

ly R

evie

w, F

iji I

slan

ds;

Bur

eau

of S

tatis

tics

Key

Sta

tistic

s; R

eser

ve B

ank

of V

anua

tu Q

uart

erly

Eco

nom

ic R

evie

w;

Cen

tral

Ban

k of

Sol

omon

Isl

ands

Qua

rter

lyR

evie

w; P

acif

ic I

slan

d E

cono

mic

Rep

orts

; Int

erna

tiona

l Mon

etar

y Fu

nd.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM9

Remittances in the Pacific

10

In 1998–99, remittances to Tonga amounted to P$63 million, more thanthree times the value of exports and very much the largest component of theeconomy; since then they have increased by about 50%. Data from Samoareveal a very similar situation. The recent growth in Fiji is remarkable andrepresents both domestic economic problems and still rapid but relativelyrecent overseas migration. Current reports indicate that remittances in Fijihave now reached US$140 million and are still rising. This is a function ofthe declining national economy, rising numbers (more than 2000) in the Britisharmy, a rise in the number of nurses and caregivers overseas, and a newmigration into security services in Iraq (Radio New Zealand International 13October 2004). In Kiribati too remittances from seafarers grew almost tenfoldbetween 1979 and 1998 (Borovnik 2004) though national data suggest thatthere was a parallel decline in remittances from Nauru. The situation maywell have been similar in Tuvalu. What little macro evidence there is suggeststhat these kinds of levels, proportions, and rates of growth are also broadlysimilar elsewhere but are clearly larger in Samoa and Tonga than in other,partially remittance-dependent economies such as Cook Islands, Kiribati,and Tuvalu.

In countries where remittances are of great national significance theyare, not surprisingly, important throughout the country. For example, a seriesof studies in Tonga in the 1980s and 1990s recorded that remittancesconstituted anywhere between 15% (a very low figure) and over 50% of theincome of various villages (Hardaker et al 1987; Faeamani 1995; Sudo 1997;Evans 2001). Similarly, in Fusi, Samoa some 90% of households receivedremittances (Muliaina 2001) while 86% did so in Nukunuku, Tonga(Halatuituia 2001) though both villages had a range of other sources of incomeincluding nearby urban employment. In more remote villages such as Ha’anoin Tonga, almost all households received remittances (Evans 2001).

Fewer households receive remittances in countries like Kiribati where arelatively small proportion of the national population is overseas at any onetime. Though the volume may not be large, the limited availability ofalternative income sources means that demands on recipients may be great.In Kiribati almost all remittance recipients gave money to those who askedfor it, and although there was an expectation of reciprocity, some recipientswere actually worse off financially by the time their remittances arrived (Clark2004). While this situation is unusual and probably rare outside Kiribati, itindicates the extreme dependence on remittances and the limited contributionthey may actually make to development.

A large number of mainly anthropological studies of one or two villagesover brief periods of time have documented how migrants’ remittancesmaintain social ties and act as insurance schemes. They are principally used,

Pacific-Remittances.pmd 15/03/2005, 10:56 AM10

The Importance of Remittances

11

at least initially, to repay debts, to finance migration of kin, and to purchaseconsumer goods including housing. They often reinforce a traditional set ofvalues that emphasizes the prevailing social hierarchy and strengthensestablished social organization. Remittances tend to go to senior familymembers who use them in traditional ways instead of for structural changessuch as land tenure reform. Overseas migrants invariably retain land rightseven after long periods of absence and may also discourage spendingremittances in nontraditional ways.

Through these social ties, remittances play a critical role within extendedPacific kinship groups. The transnational corporation of kin not only helpsto maintain family and communal networks but may even enlarge networksof support and empowerment. Despite occasional conflicts within householdsabout economic and social strategies, migrants might also be seen as actingas a family firm in terms of both consumption and investment in differentgeographical locations (Bertram 1999). Poirine has suggested that remittancesmay even be seen as informal loan agreements (1997; Brown and Poirine 2005).

Marcus emphasized over a decade ago that, “…the capacity to call oninternational resources has become a crucial factor in influencing a family’slocal economic conditions. The lowest stratum in contemporary Tonga are[sic] those totally dependent on the nation state framework, and the limitedresources it embodies, without any overseas options at all” (1993). It isincreasingly argued as in Tonga that, “…every family needs to have someoneoverseas. Otherwise the family is to be pitied” (quoted in Small 1997). Hence,in contrast to western societies, it is often female-headed households thatsurvive most effectively (Gailey 1992a).

Remittances have raised living standards in the islands (Faeamani 1995),have contributed to employment especially in the service and constructionsectors, and have eased balance of payment problems (despite contributingto inflation) especially in the larger Polynesian countries. However, increaseddemand for improved consumer goods can usually only be met by furthermigration, and it has generally been argued that little of the remittance incomehas been invested in economic growth.

In some countries, international migration has been viewed as a kind ofsafety valve reducing pressures on governments to provide employment andwelfare especially when the rate of population growth is high and that ofeconomic growth is low. Thus in virtually all the home countries in the Pacific,there is little concern over international migration though there are concernsabout specific issues such as absolute population loss, brain drain, and theuse of remittances for development. Nevertheless, none of these concernshas ever been translated into policy. Rather, throughout the region, the safety-valve effect, limited economic growth, and priorities given to freedom of

Pacific-Remittances.pmd 15/03/2005, 10:56 AM11

Remittances in the Pacific

12

movement have resulted in steady, unimpeded migration though bonds havebeen imposed on many of those receiving higher education abroad to reducethe skill drain. In both Samoa and Tonga, the size of the domestic populationhas therefore remained unchanged for much more than a decade despiterelatively high rates of natural increase.

In Tuvalu, where international migration to Nauru declined and reversedfollowing the closure of the phosphate mine and where there have beenconcerns over rising sea levels, there has been pressure on Australia since1994 to increase migration opportunities. The potential impact of thegreenhouse effect in other atoll states has also resulted in increased interestin international migration (Connell and Lea 1992; Connell 1999, 2003a). InFSM and the Marshall Islands, the possibility of migration to the UnitedStates was written into the Compact of Free Association when the two statesmoved towards independence (Connell 1986). The possibility of restrictionson international migration, on the other hand, has discouraged independencein the Cook Islands, Wallis and Futuna, and American Samoa.

There is little evidence so far that the safety valve of internationalmigration has enabled home countries to significantly reduce their populationsand/or to use remittances to restructure their economies. Migration has tendedto be viewed rather as a substitute for development or even as developmentitself instead of as short term support for national efforts. The use ofremittances is therefore of greater significance.

Overall, the available evidence on international migration in the Pacificislands demonstrates that in the short run a number of distinct benefits accruesto individual migrants and their families and to home countries. Despite risingunemployment and recession in host countries, this appears to remain truealthough there are many poor migrant families. Migration has reduced thelevel of open and disguised unemployment although it has also contributedto a loss of skilled human resources. In the case of migration of Samoans andTongans to the United States alone, “Emigration results in the permanentloss of young educated skilled labor from the Pacific island nations. Skilledlabor is in short supply and emigration probably hinders development”(Ahlburg and Levin 1990). This is certainly true in the health sector wheremore costly (and less skilled) replacements have been required (Brown andConnell 2004a; Connell 2004) and is widely true in the government sector inSamoa (Liki 1994) and almost certainly true elsewhere. Exceptions occurwhere wages and salary levels are more comparable with those in the mainmigrant destinations. Thus in Cook Islands, the extensive migration of skilledpersonnel is balanced by a considerable degree of return migration (Hookerand Varcoe 1999).

Pacific-Remittances.pmd 15/03/2005, 10:56 AM12

The Importance of Remittances

13

Despite the benefits of international migration, there are indications thatin the long term it may impose considerable costs. Governments have notbeen able to control or to direct the use of remittances nor have they generallysought to do so when migration generates increased demand for consumergoods. This demand can usually only be satisfied by further migration ifother sources of national income are difficult to develop. Inequalities mayincrease, and establishing population and migration policies may bepostponed.

The available data on remittances in the Pacific are fragmented andlimited, hence many questions have not been thoroughly investigated. Thoseinclude the extent to which remittances are sensitive to variables like foreignexchange rates, relative rates of interest and inflation and the extent to whichmigration and remittances contribute to inequality. Are demographiccharacteristics (age, gender, duration of stay, residential status), economiccharacteristics (employment status, income levels, etc.), and conditions ofmigrants in home countries good predictors of remittance levels? How dofamily reunification and the potential for returning home affect propensitiesto remit? In the absence of adequate data, answers may be inaccurate or mayapply only to specific locations and times.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM13

Remittances in the Pacific

14

Since remittances take many forms (including goods and commodities)and pass through many channels, there are obstacles to definitively assessingtheir value. Recorded remittances are sometimes a poor measure anddefinitely an underestimate of the actual total. Much of the analysis ofmigration and remittances has been based on two suppositions. The first isthat remittances pass through the official banking system. These are mostoften recorded in banking statistics and in the home country’s balance ofpayments current account as “unrequited transfers.” Conclusions about themagnitudes and long-term trends in remittance flows are based on such dataeven though it is well known that a considerable and possibly increasingproportion of remittances enters through other channels and in differentforms (Ahlburg 1991). In recent years, Western Union and ATMs have beendestinations for remittances, but how significant electronic transfers havebecome is unknown.

A very substantial amount of cash is sent informally with people travellingto the Pacific islands especially during the Christmas period. Thus, the overallflow of income into the islands is substantially higher than most formalestimates suggest and is not noted in official records of capital transfers.Brown (1995) has estimated that unrecorded remittances from Australiaamounted to as much as 34–41% of the total from Tongans and 42–60%from Samoans. The way money is sent varies according to the purpose, theamount, the urgency, and the availability of a reliable carrier. In the case ofTongans in Auckland, about 65% claimed to have sent cash through formalchannels while smaller sums for subsistence needs were often enclosed inletters (Fuka 1985). In Fusi, Samoa, remittances were commonly sent inletters in the past but are now more likely to be hand carried by relativesbecause of numerous thefts from the postal system (Muliaina 2001).

The use of channels other than the banking system seems to be partly aresult of underdeveloped national financial systems in Samoa and Tonga andin remote regions. Charges on remittances have also resulted in disincentivesto use banks and have encouraged hand carrying. In both Samoa and Tonga,foreign exchange agents operate openly (Ahlburg 1991). Though the informalrate of exchange is rarely very different from the official rate, this system isvery convenient.

The Composition of Remittances3

Pacific-Remittances.pmd 15/03/2005, 10:56 AM14

The Composition of Remittances

15

The second supposition is that remittances are usually cash, postal orders,or checks.

Actually, most also include various kinds of goods. Early movements,especially in Melanesia, found migrants returning almost entirely with goodsbecause of the limited utility (and low prestige) of money in many remoteareas and the fact that they received goods as wages. Remittances from earlymigrants from Polynesia to New Zealand were also mainly goods. In the late1970s some 60% of remittances from Polynesians in Wellington were gifts,goods, and also services. Consumer items were (and are) more costly inPacific island states, hence it was rational to convert earnings into goods.Subsequently cash became more important, but goods have remainedsignificant, especially in Tonga. Thus in the early 1980s, remittances fromTongans in Auckland were 76% cash and 24% goods (Fuka 1985), and in theSamoan village of Vaega, goods constituted 40% of remittances in the 1980s(O’Meara 1990). In both cases, most of the goods were food and second-hand clothes, but household items such as blankets, sheets, and utensils werealso sent. Other estimates of the proportion of remittances in kind are similar(Ahlburg 1991) although there has again been an increase in recent years.

From the mid-1980s, there was an important transition in Nuku’alofa,the capital of Tonga, as goods that were initially sent for individual householduse became the basis of a very substantial flea market and were thustransformed into investments (Brown and Connell 1993a). Cars, which oftenbecame taxis, were also remitted to Tonga (Ahlburg 1991) especially fromAmerica. In 1999 in the village of Nukunuku not far from Nuku’alofa, nomore than 34% of all remittances was cash; the remainder was goodsespecially food though vehicles represented 16% of the total (Halatuituia2001). In the village of Fusi, which is about the same distance from Apia, thecapital of Samoa, there was a similar transition from cash to goods; mostwere luxury items, but food and second-hand clothes were also common.Food mainly came from nearby American Samoa, and other goods camefrom New Zealand. There has also been a significant increase in the numberof street sellers in Apia, though the growth of market trading has been muchmore substantial in Nuku’alofa. Recently, two thirds of Samoan remitters inNew Zealand sent only goods (Muliaina 2001) while two thirds of remittancesfrom Tuvaluans were in cash mainly because of the very considerable expenseand difficulty in getting goods to the islands, though migrants would havepreferred to send them (Simati and Gibson 2001).

In almost every case, remittances flow in both directions. Those sentfrom home countries are most likely to be food and handicrafts, and little isknown about them. In some cases they are examples of traditional reciprocity;in others they are an attempt to elicit more remittances. Other than in the

Pacific-Remittances.pmd 15/03/2005, 10:56 AM15

Remittances in the Pacific

16

exceptional case of parts of formerly American Micronesia, the formal marketvalue of remittances from home is rarely comparable to that from migrants.Alexeyeff describes them as, “…a micro-example of the economic dependencythat is seen to characterise the Cook Islands diaspora” (2004). In reality,however, these offerings have extraordinary symbolic and nostalgic value(Alexeyeff 2004). In Rotuma, Fiji Islands and almost certainly elsewhere,migrants preferred to receive goods from home since they represented timeand effort and signified caring in a tangible form (Rensel 1993). SomeSamoans in Auckland even claimed that they sent remittances in order toreceive traditional goods including food (Muliaina 2001). Tongan womenhave sent substantial quantities of traditional wealth (koloa), mainlyhandicrafts, to migrant communities; in return migrant communities providehospitality and significant cash payments (James 1997; Small 1995, 1997).Though the role of women in the production of handicraft goods for exchangeis relegated to “the economy of affection,” it is nevertheless an importantcomponent of small business enterprises and other commercial transactions(James 1997).

Groups from several countries, including Cook Islands and Samoa,regularly make travelling visits (malaga) to relatives overseas taking kava,mats, and food items. Rotumans (Rensel 1993) and other Fijians (Stanwixand Connell 1995; Scott 2003) also travel to migrant communities overseasto generate remittances or cash donations for domestic activities. Thesetravelling cultural groups often give musical performances for financial benefit(James 1993). In recent years, the increased volume of goods remitted fromSamoa and Tonga may represent both payments for goods sent from hostcountries and exports to be sold in host markets (Besnier 2004; Brown andConnell 1993). The extent to which remittances in kind from the Pacificisland states are merely symbolic transfers, are stimuli for further remittances,or are part of an import-export system varies and is largely unknown.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM16

The Importance of Remittances

17

Though there have been many studies of the rationale for migration,there have been virtually no studies on the specific rationale for sendingremittances. Decisions to migrate are often made by families and rarely byindividuals. Migrants leave to meet certain family expectations the key oneof which is remittances. Indeed to some extent there is an implicit contractthat they will be sent, and the costs of not doing so may be high. Futurefamily solidarity; rights to inherit family assets, especially land; and prestigein the village community may be at stake (Ahlburg 1991; Macpherson 1994).

There is a general expectation that migration will benefit extendedhouseholds and sometimes even whole clans or villages as it did in the earlyyears of migration in Papua New Guinea and that it may continue to do sothrough support for village organizations. Overall, migration is expected toimprove both the living standards of those who remain at home and the lifestyles and incomes of the migrants. More specifically, in rural areas andhome islands as in Tonga, “…migration and remittances …are perceivedsolely as a means for improving family incomes and welfare rather than fordirect or indirect national economic development” (Morton Lee 2004;Tongamoa 1987). In Samoa migrating was simply, “Cto seek wealth for all”(Muliaina 2001).

The underlying economic rationale for migration and for its benefitsextends beyond the households in the Pacific islands so that the,“...international scale of family operations, still tied to kin at home… wouldequal or exceed in value the Tongan national product” (Marcus 1981).Developing this notion, Bertram has suggested that Tongan and other similarhouseholds are characterized, “…by remittance transfers among variouscomponent parts of the transnational corporations of kin which direct theallocation of each island’s family labour around the regional economy” (1986).Consequently, “…families deliberate carefully about which members wouldbe most likely to do well overseas and be reliable in sending remittances”(Gailey 1992b; Cowling 1990).

Not surprisingly, the significance of remittances at the household andnational levels has led to concern over the possibility of future declines dueto changing migration rates, recessions in host states, or other factors (Mileset al. 1992; Marcus 1993; James 1991; Campbell 1992). Though remittances

The Rationale for Remittances4

Pacific-Remittances.pmd 15/03/2005, 10:56 AM17

Remittances in the Pacific

18

are driven by family and community needs and to some extent by the savingsand investment objectives of migrants and their families, they are stronglyaffected by income levels in host countries and are thus vulnerable torecessions. There is little that can be done about such fluctuations except tobe aware of their existence and to make some provision for their occurrence.One strategy is to attract and retain migrant savings to invest in businesses insuch periods.

The most detailed accounts of the intended purposes of remittances(which were not necessarily the ultimate uses) were conducted among Tongansin Auckland and Sydney (Fuka 1985; Tongamoa 1987). Both these studiesfound four principal intended uses for cash. The first was for subsistencewhich covered daily grocery needs and minor church, school, village andfamily obligations. Some 79% of remitters in Auckland and about 85% inSydney sent money for these expenses. Subsistence money was especiallyimportant in households of elderly couples and teenage girls with no one togrow food crops. These basic needs have become less significant over time.

The second purpose was for church donations. In Tonga, regular donationsare expected from and allocated to church members, and substantial donationsbring considerable respect and status. This category is likely, therefore, to besignificantly higher for Tongans than for other migrant groups. Approximatelyhalf of Tongan remitters in Auckland and 76% in Sydney sent remittancesfor church donations. The third intended purpose was for school fees matchingthe significance attached to formal education in the Pacific, and the fourthwas for major family occasions such as funerals, birthdays, and weddings.Such remittances are obviously intermittent although they are often large.Beyond these four principal objectives, there were a variety of other intentionsincluding the building or purchase of houses, the establishment of a localbusiness, or the purchase of a vehicle.

Samoans in Auckland saw themselves as having a series of remittancetargets to achieve: repayment of a family loan, building a family house,contributing to a wedding, bestowing chiefly titles, and contributing to funeralexpenses (Muliaina 2001). Purchasing airfares is also a common use. Buildinga European-style house has always been a key goal of most migrants since itis a symbol of achievement. Many apparently permanent migrants remitmoney for the construction of permanent houses for themselves that are usedby kin or rented out in their absence. Migrants also frequently send remittancesto cover the needs that follow crises. Cyclones have been the single mostimportant trigger as was evident in Samoa in 1991 and 1992 (Muliaina 2001),after Cyclone Heta in Niue in 2004, and in Tonga in 1982 and 1983 after thewidespread damage caused by Cyclone Isaac (Fuka 1985).

Pacific-Remittances.pmd 15/03/2005, 10:56 AM18

The Rationale for Remittances

19

Remittances for investment in or purchase of local businesses havebecome increasingly significant especially for corner stores, for fishing(including boats and gear), and for taxis. Those who sent remittances to investin shops stated that this was a means of ensuring that the recipients had aready supply of cash. In due course they may also benefit the remitter if he orshe returns. Although both shops and vehicles absorbed a very smallproportion of all remittances sent from Auckland and were non-existent inSydney in the early and mid-1980s (Fuka 1985; Tongamoa 1987), in lateryears their significance grew considerably (Faeamani 1995; Small 1997) asconsumption shifted to investment. Moreover while Samoan migrants believedthat their remittances were used mainly for consumption (Walker and Brown1995), in reality a significant proportion was used for investment as recipientswere anxious for long-term benefits.

Migrants have also contributed to various collective endeavors in responseto visiting groups’ fund-raising activities such as dances organized by footballteams or home villages to generate funds for equipment, water systems, orchurch roofs (Fuka 1985). In this institutional context there is probably aparticularly high correlation between the intention of the remitters and theeventual use of the remittances.

Over the years, remitters have been more likely to respond to specificrequests from home. Senders’ intentions may therefore increasingly matchthose of the recipients. About 63% of Tongans in Auckland remitting in 1984claimed to send money or goods only when kin in the islands requested it; incontrast 20% set aside a regular sum, and the remainder stated that they didboth. Those who responded only to requests had their families mainly inAuckland whereas those who regularly set aside money had most of theirfamilies in Tonga (Fuka 1985). On balance, the evidence suggests thatremitters respond rather than initiate and that the probability of migrantsinitiating rapidly declines over time.

The intentions of Tongan remitters in Sydney correlated closely with theuses specified by Tongans on the main island of Tongatapu. Remitters intendedmore to be used for ceremonials, and recipients intended to use more foragricultural development once again demonstrating recipients’ interest in long-term benefits, but otherwise intentions and reality were very closely linked(Tongamoa 1987). Similarly, Tongans on Ha’ano requested remittances forspecific purposes and senders responded rather than sending regular sumswithout a specific purpose (Evans 2001; Muliaina 2001). Both these casessuggest that other sources of income are not diverted from their usual purposeby remittance flows.

In the case of Cook Islanders in New Zealand, remittances have notsimply been an automatic response to customary obligations, material

Pacific-Remittances.pmd 15/03/2005, 10:56 AM19

Remittances in the Pacific

20

circumstances, or ceremonial requirements. Remitters have consciouslyweighed their own financial situations against ongoing (and new) obligations,life crises, and the changing circumstances of their relatives in the islands.Overseas migrants often maintain contact by letter, telephone, and visitingfriends and are knowledgeable enough about the material circumstances ofisland kin to change their remitting practices to meet their perceptions oftheir kin’s needs (Loomis 1990; Underhill, 1989). Thus migrant intentionshave largely been determined by the needs of those at home. Although thereappears to be a transition towards remittances on demand, again much moreneeds to be known before such a conclusion can be reached.

Finally, migrants actually transfer income for their own later use to financeretirement or to acquire assets such as real estate. In some instances, incomeis accumulated overseas and transferred as a lump sum at the time of themigrant’s return (Ahlburg and Brown 1997). Whether such transfers canactually be defined as remittances is doubtful.

Perhaps surprisingly, but also significantly, some two-thirds of Tongansin Auckland were indifferent to how their remittances were actually used,and less than 1% was concerned about their specific use. Of this very smallgroup, some complained that the use was different from what they had intendedwhile others feared permanent demands for remittances. Only in Port Vila,Vanuatu, were strong sentiments expressed over the intended use of remittances.Migrants from outer islands preferred to send food rather than cash unless itwas needed for something such as school fees or wedding expenses(Haberkorn 1989). Similarly skilled workers in Honiara, Solomon Islands,were more likely to contribute to rural education expenses (Bellam 1964).

Such statements and attitudes often indicate migrants’ feelings that kinat home waste some of the remittances they are sent whereas other evidencesuggests they are more likely to invest it. In the mid-1990s, Tongans in Americacomplained that village people were eating remittances sent for public projectsby redirecting the money to private consumption (Small 1997). Samoans inNew Zealand complained of mismanagement and corruption (Macpherson,1994) and for Tongans in Melbourne there was an, “…undercurrent ofresentment for both receivers and remitters” (Morton Lee 2004). For the vastmajority, however, doing one’s duty was sufficient in and of itself and wasthe basis for the migration-remittances system. At the same time, doing one’sduty disguises and inflates very valid reasons for maintaining contacts withhome areas: remittances are a powerful element of social insurance forpossible return.

Unfortunately, there are not enough good studies of the intended andactual use of remittances or their impact on other expenditures to draw parallelsbetween the detailed case of overseas Tongans and groups elsewhere.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM20

The Rationale for Remittances

21

However, the correlation between intentions and actuality in these cases andthe similarity between the intended and actual use in various circumstancessuggests that these cases are fairly typical.

The proportion of income remitted by international migrants is usuallyhigh. In Auckland in 1984, Tongans were remitting on average some 12.8%of their net though the range was considerable (Fuka 1995). In Sydney in1986, the amounts sent by Tongan migrants also varied substantially andwere sometimes more than 50% of their incomes (Tongamoa 1987). Thesefigures are unusually high, and suggest that migrants may have exaggeratedtheir capacity to remit. Using survey data for Samoa and Tonga from boththe remitting and receiving ends, it has been estimated that the average Tonganmigrant in 1994 remitted US$2,043 per annum, and the average Samoanmigrant remitted US$789. These represented 13.1 and 5.3% of averagemigrant income respectively (Brown 1995). Disproportionately fewerTuvaluans in New Zealand remitted at all, and the average totals were muchlower. Tuvaluans were, however, mainly concentrated in very low-payingjobs and had spent little time in the country (Simati and Gibson 2001).

Most studies of remittances have pointed to the sacrifices that migrantsmake and to the goods they forgo to support distant extended families. ForSamoans,

As a result of the remittance economy immigrants to NewZealand may find it difficult to save the funds necessary formortgages, hire-purchases and other financial transactions. Helentold me, ‘You have this wonderful budget that works until someonedies the next week.’ (Wurtzburg 2004)

For Tonga, it has been said that, “…overseas Tongans who continue toplay the patron role to their relatives often bankrupt themselves in the processand to avoid this must either give to fewer relatives, give more intermittentlyor give less than is needed or wanted” (Small 1997).

In the particular case of migrants from Kiribati and Tuvalu working oninternational shipping lines, the amount of money to be remitted is negotiatedbetween the seafarer and his dependents prior to departure and is written intothe contract. In most cases this is usually around 70–80% of earnings leavingthe rest for the seafarer to spend or save during his contract. Those highpercentages are not always reached, however, because of deductions for over-expenditure on personal interests (often alcohol and the company of otherwomen) and because of fluctuating exchange rates (Clark 2004). Seafarersalso sometimes take advances on wages and spend money at stopovers onthe way home thus decreasing the total (Dennis 2003).

Pacific-Remittances.pmd 15/03/2005, 10:56 AM21

Remittances in the Pacific

22

In many cases individuals and households transfer remittances betweenorganizations or to organizations in the home country. In these transactions,expectations are very likely to match outcomes. This institutional remittinghas been well described for Tongans in Sydney. It is particularly evident inreligious donations, for football teams of various kinds, for brass bands, forwomen’s groups, and for various development projects. Fundraising is carriedout at social functions such as concerts, kava parties, and dances (Tongamoa1987; Connell 1991; Macpherson 1994). Even migrants who do not otherwisesend money or goods tend to contribute to these collective efforts (MortonLee 2004). Although a substantial amount of money and goods flows throughinstitutional channels, the less regular nature of these transfers has meantthat there are few if any estimates of their significance.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM22

The Rationale for Remittances

23

Almost every study of remittances and their rural-urban and internationallinks suggests they will decline over time although few studies have evaluatedto what extent or over what time period this might occur (Connell 1991).More recent studies have, however, shown that this generalization may notnecessarily be true and that the period of sending remittances may actuallybe very long. Many studies have also not taken into account the significanceof new electronic media that have transformed the nature of links and perhapsof remittance practices.

A decline in the level of remittances over time would nonetheless scarcelybe surprising. Social ties and distant perceptions of needs and wants are likelyto fade, successful migrants may be followed by others from the same family,initial targets (where they existed) will have been met, and local investmentand expenditure will be more rewarding and necessary as the probability ofreturn declines. Although migrants face a life cycle of obligations to theirhome areas, those obligations are likely to lose their immediacy, to competewith new obligations, and to be increasingly ignored (Fuka 1985).

Remittances are sustained over longer periods when the probability ofreturn is greatest. The highest proportions of incomes remitted by Tongansin New Zealand were from those who had been there for fewer than 4 years,indicating that specific goals were expected of migrants and that new migrantswere more likely to perceive migration as temporary (Fuka 1985). Similarly,in the village of Sa’asi in Samoa, temporary migrants comprised only 16%of the village’s remitters but supplied almost one third of all remittanceincome.

Temporary migrants remitted more partly because many of their expenseswere met by permanent migrants and partly because their temporary visasensured that their return was imminent. Migrants permanently overseas wereunder less immediate pressure to remit, and they had also acquired financialcommitments in New Zealand, hence their village commitments were lessintense and less significant (Shankman 1976).

In one group of Samoans in Auckland, two thirds had decreased theirremittances over time and a third had increased them (though the time periodwas not stated). Those who had decreased their remittances said the reasonsincluded a reduction in take-home pay, increases in the cost of living, the

The Duration of Remittances5

Pacific-Remittances.pmd 15/03/2005, 10:56 AM23

Remittances in the Pacific

24

birth of additional children, the death of parents or close relatives on thehome island, family reunification, and a change in marital status. Several ofthose who had remitted substantial sums as single women now saw theirspouses and children absorbing their incomes. In contrast, those who increasedtheir remittances saw this as responding to a greater number of requests fromSamoa. Furthermore, if the remitter’s children were adults, their demandswere fewer so they could add to their remittances. For the whole of thisgroup, some 95% were emphatic that they would continue to send remittancesfor the rest of their lives though they recognized that their children werealready questioning their personal gains from the practice (Muliaiana 2001).In the context of long-term household migration such a pattern now appearsrelatively typical.

Two relatively detailed studies of remittance flows over time wereconducted on overseas Tongans. The first was in Auckland and concludedthat after 15 years, non-remitters became as numerous as remitters and thatafter 20 years, more than two thirds no longer remitted (Fuka 1985). However,the results are more complicated in that remittances tended to drop after 4years of residence in New Zealand but rose again after a decade while thosewho were still remitting after 15-19 years actually sent the greatest amounts.However, the proportion of income sent showed a slightly more consistentdecline from 19.6% for a residence period of fewer than 5 years to 12.9%after 20 years (Fuka 1985). The same pattern was found among Tongans inSydney: the amount increased up to around 7 years but then began to declinealthough migrants who had been in Australia for more than 18 years stillremitted (Tongamoa 1987).

Thus, for Tongans overseas, remittances did decline over time, but themanner in which this occurred, for whom, and why, is far from clear. Abroadly similar pattern was reported for Tuvaluans in New Zealand whereremittances were initially low (as migrants sought to get established andinitially had low-paying jobs) and rose after 4 years. They appear to havebeen sustained for as long as 30 years (Simati and Gibson 2001) though fewTuvaluans have been in New Zealand that long. What was striking and waswell documented for Cook Islanders and for Tongans (Loomis 1990) wasthe considerable time and high level at which remittances were maintainedwith only slight evidence of the anticipated decay though the number ofmigrants studied was too small to be statistically reliable. Nonetheless allsuggest that time itself is certainly not the key variable.

These studies and others suggest that the two key variables that doinfluence the sustainability of remittances are permanent residence status ofremitters in the host country and whether recipients had joined their relativesin the host country or had died. Where these events have occurred remittances

Pacific-Remittances.pmd 15/03/2005, 10:56 AM24

The Duration of Remittances

25

are virtually non-existent (Fuka 1995; Grieco 2003, 2004). In the case ofTuvaluans in New Zealand, not only was the presence of at least one parentin Tuvalu the principal factor influencing remittances, but remittancesincreased as family size in New Zealand increased (Simati and Gibson 2001)demonstrating both that households with higher disposable incomes remittedmore and that remittances increased with the ability to remit. Generally, asin the case of Tongans in New Zealand and Fijians in Sydney (Fuka 1985;Stanwix and Connell 1995), the volume and regularity of remittances is afunction of intent to return, though this seemed not to be the case for Tuvaluans(Simati and Gibson 2001).

Few studies of remittances have had large enough samples to bestatistically valid. One of the only two large surveys of migrants from Pacificisland states was carried out in Australia in the mid-1990s. Econometricanalysis of cross-sectional data from a survey of Samoan and Tongan migrantsin Brisbane and Sydney found that while the propensity to remit was negativelyrelated to the age of the migrant, it was positively related to the migrant’slength of absence from home (Walker and Brown 1995). More importantlythere was little evidence of remittance decay to the extent that, “…none ofthe assumptions about migrants’ remittance behaviour on which the doomsday,remittance-decay scenario is based, is valid [and] there is also much lesscause for pessimism concerning the sustainability of remittance levels”(Brown 1998). As long as there was no change in the size of the migrantcommunity, the aggregate level of remittances did not fall. In other words,migrants were highly likely to continue to meet the needs of those at home.

In contrast, Grieco’s analysis of survey data on Micronesians (from FSM)in Guam and Hawaii demonstrates that there was remittance decay in termsof both the probability of remitting and the volume remitted. Remittancevolumes were strongly correlated with household incomes and with familyunification in the host country; households that participated heavily in migrantsocial networks and were less assimilated as measured by language choicewere most likely to remit while those that had reunited in the host country orwhose relatives at home had died were least likely (Grieco 2003). Theconclusions were similar for Samoans and Tongans in Australia.

A further econometric analysis of the data on Samoan and Tongan migranthouseholds in Australia in which one of the members was a nursedemonstrated that they sustained remittances at a very high level over a longperiod of time. Indeed a significantly higher proportion of householdscontaining nurses both sent remittances and remitted more than averagehouseholds while remittances from households without nurses showed atendency to decline with time when all other characteristics were constant(Brown and Connell 2004). On the other hand, households with nurses were

Pacific-Remittances.pmd 15/03/2005, 10:56 AM25

Remittances in the Pacific

26

ultimately less generous than other households in terms of the proportion ofincome that they remitted though the amounts were in excess of the cost oftheir training, and the proportion fell steadily as their income increased(Connell and Brown 2004). This suggests that after achieving a target,remittances no longer rise. There is no other data that disaggregatesremittances by occupational group other than for seamen in Kiribati thoughthere are various references to the inability of the most poorly paid to remit(Grieco 2003, Simati and Gibson 2001). Consequently the extent to whichnurses are distinct from other groups is not clear.

Most remittances have primarily been given to parents or children andsecondarily to brothers and sisters (Tongamoa 1987; James 1991; Fuka 1985;Rensel 1993). However in the distinct case of seafarers from Kiribati andTuvalu, all of whom are male contract workers, a higher proportion goes towives than to parents (Clark 2004; Borovnik 2003, 2004; Dennis 2003). Wheremigrants have children on the home island, as is often the case, thenremittances are likely to be both substantial and sustained (Underhill 1989).Again, in the case of Tongans in New Zealand, those who remitted the mosthad close relatives in Tonga—wives, children, and parents. For Tongans inSydney, parents were the major recipients, but brothers and sisters alsoreceived substantial sums whereas other kin received relatively little(Tongamoa 1987). In one Tongan village in Ha’apai, over 60% of remittanceswent from children to their parents and 10% went from husbands to theirwives (Evans 1996; 2001). This kind of situation has been particularly welldescribed in Tonga where the ideology of sacrifice and service due to parentsis supported both by the traditional structure of Tongan culture and byChristian teaching (James 1997). Much the same is true in Samoa whereMacpherson emphasizes that remittances are likely to dry up when parentsdie or join their kin overseas (1994). Remitting to the nuclear family isprobably even more evident elsewhere, especially outside Polynesia wherethe volume is smaller.

The location of household members is thus highly critical for the flow ofremittances. For Tongans in New Zealand there was a correlation betweenthe number of dependents in Tonga and the size of remittances (Fuka 1995).Correspondingly, the larger the number of dependents in the host country,the less likely the migrant was to remit. Those who sent little from Aucklandtended to have their parents in New Zealand, and those who sent nothing atall tended to comment, “We no longer remit because none of us are in Tonga.”Similarly, while remittances to parents were, … “whenever I have some sparemoney,” those to siblings and grandparents were, … “when they write andask” (Fuka 1985). While this conflicts with the Tuvaluan case, the differencemay be a function of overall time spent in New Zealand.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM26

The Duration of Remittances

27

Younger overseas residents were less likely to remit. Amongst Tongansin Melbourne, very few under the age of 30 sent money or goods directly toTonga, and even older migrants said that they sent goods only infrequentlyand intermittently. Few migrants sent remittances regularly; most preferredto wait until they received requests for money or goods for particular purposeswhile those who sent nothing stated that they had no family left in Tonga(Morton Lee 2004). Samoans in Auckland were even more blunt. None ofthe adult children who had been born in New Zealand or who had arrived assmall children had been asked directly for remittances, but their parents hadasked them to send some. Though they had not declined the requests, they,“...all admitted to the feeling of resentment to the remittance practice andsaid they would like to escape this obligation.” Sixty percent of them hadvisited Samoa and believed there were development opportunities there,“…in the form of land and marine resources which their relatives appearednot to have noticed,” and that too much of their remittances went intoconspicuous consumption (Muliaina 2001). Tongans in Sydney had somewhatsimilar perspectives.

Generally in Tonga, it has been argued that the lack of close relatives athome is more likely to be the principal reason for low remittance levels ratherthan any loss of traditional customs among the second generation (Fuka,1985). The limited available evidence on the remittance patterns of secondgenerations indicates that, as Morton Lee (2003, 2004a,b) and Muliaina (2001)have emphasized, they respond only indirectly through the urgings of theirparents and their sense of family, and they therefore contribute very limitedsums. This is particularly significant as migration opportunities decline andthe number of overseas born “islanders” becomes the majority. Thus overseas-born Tongans in New Zealand and in Melbourne (Morton Lee 2004b) alongwith Cook Islanders, Niueans, and others in New Zealand are now a majorityrather than a minority. Not only does this probably mean that for all thesegroups their remittances are limited, but their social and economic ties arelikely to increasingly be with each other rather than with home. Morton Leehas thus concluded that,

Unless there are profound changes in the relationship of theyounger generations with the Tongan ‘homeland’ and in their senseof ‘belonging,’ the prospect of maintaining current levels ofremittances is remote, which gives serious concern for Tonga’seconomic situation (2004b).

She further warned against the complacency of many people in Tongaand of institutions outside that remittances will continue into the indefinite

Pacific-Remittances.pmd 15/03/2005, 10:56 AM27

Remittances in the Pacific

28

future (ibid). As overseas generations lose language and cultural skills, theirsense of belonging must decline, and furthermore, there is no certainty thatmigrants’ economic status will always improve.

The probability of remitting is related to attitudes to the needs of kinremaining at home, the number and location of kin and dependents, the attitudeof migrants to overseas commitments, and the ability—in terms ofemployment status and income—to send remittances. It is apparent that, insome cases, migrants may respond so strongly to the perceived need forremittances that they defer purchasing a house as appears to have been thecase for Tongans in the northern suburbs of Sydney (Faiva 1989). A range ofhousehold goods may also be foregone (Fuka 1995) and even retirementmay be postponed (Muliaina 2001) to the extent that there is some degree ofsacrifice. Indeed, 28% of Samoans in Auckland claimed that they hadexperienced financial stress because of remittances though none had everavoided them. Moreover, all second generation Samoans in Auckland claimedthat on at least one occasion meeting remittance requests had put them inarrears with mortgage payments (Muliaina 2001).

One of the inevitable outcomes of migration is marriage outside the ethnicgroup; around half of all marriages of Tongans in Australia involve partnerswho are not Tongan. In such circumstances the probability of sendingremittances declines significantly as such couples are more likely to purchasehomes and save money (Morton Lee 2004a, 2004b).

Gender plays an important role in remittances. Virtually without exception(where there are data) women tend to be more frequent remitters althoughthey may lack the earning capacity to send the same amounts as men. Samoanand Tuvaluan women in New Zealand were specifically referred to as themost reliable remitters (Shankman 1976; Simati and Gibson 2001). AmongAuckland Tongans, nonremitters were much more likely to be males. Althoughthe average sum sent by men was higher than that of women, the latter sent15% of their net incomes compared with 11% for men (Fuka 1985). Thesame situation occurs in other Polynesian countries and has been attributedto their social structures in which women have greater responsibility formaintaining the household. Women in general have a much betterunderstanding of household needs throughout the Pacific (Stanwix andConnell 1995) and are more likely to respond to them. In Auckland it waswomen particularly who stressed that they had been selected by families tobe migrants because they would more reliably remit large proportions oftheir incomes than untitled men would (Muliaina 2001). Tongan women aremore likely to remit to sisters and daughters after the needs of parents havebeen satisfied (Gailey 1992a). Both female recipients and dispensers ofremittances take responsibility for their investment decisions and their

Pacific-Remittances.pmd 15/03/2005, 10:56 AM28

The Duration of Remittances

29

management acumen appears to be strengthened as time passes (Fuka 1985;Latailakepa 1997).

Income levels as such appear not to be a significant determinant of theprobability that migrants will remit, though more poorly paid (often more recent)migrants are less likely to do so. Though there is some relationship betweenincome levels and the amount of remittances (Walker and Brown 1995; cf.Brown 1998), other factors, particularly employment status, appear to be moreimportant. Similarly, the relationship between remittance flows and broadlydefined development or need in home areas is unknown. Though it is apparentthat remittances are often sent in response to requests, these may not necessarilyreflect need, though they are often substantial in the wake of natural disasters.

Overall, for Tongans in Auckland, the strongest influences on remittanceflows were residential status, frequency of return trips (which may reflectboth social ties and perceptions of need) and whether migrants expected toremain in New Zealand or return to Tonga (Fuka 1985). As in the case ofTongans and Samoans in Australia, there was also a correlation between theintent to return and the level of remittances, while intending returnees alsoaccumulated more financial capital at home (Ahlburg and Brown 1998). Inthis context as in most others, migrant remittances displayed components ofboth altruism and self-interest. At least for Tonga, James (1997) prefers tosee all remittances as the outcome of a “contract” rather than altruism, whileBrown and Poirine (2004) argue somewhat similarly that remittances areneither a function of pure altruism or pure self-interest but are better explainedas “impure altruism.”

The bulk of studies of long-term migration point to shifts in remittancesthat increasingly favor the interests of the remitters. Initially remittances aresent to parents which is clearly the case in the Pacific, and in an economicsense can be seen as repayment for their past investment in the human capitalof the migrant. In a social sense, this is usually expressed as duty, loyalty,and maintenance of family ties. A second wave of remittances is subsequentlymore likely to be dominated by brothers and sisters and by children; thisphase may also correspond with a decline in volume. Indeed, decline followingthe death of parents seems ubiquitous (e.g. Muliaina 2001). That phase canbe seen as an investment in the human capital of the next generation (Brownand Poirine 2004). The third and final phase represents payments to spousesand indirectly (via investment) to the remitters themselves, as return migrationand/or retirement approach. In Kiribati (and probably Tuvalu), this takes adifferent form as the actual distribution of remittances from seafarers to wivesor parents is relatively unchanged over time, but greater sums are placed inthe seafarers’ personal accounts to be used for investment, usually in housesor land, rather than for immediate spending (Borovnik 2004).

Pacific-Remittances.pmd 15/03/2005, 10:56 AM29

Remittances in the Pacific

30

The majority of studies of remittance use in the Pacific suggest that despiteconsiderable diversity they primarily are spent on consumption. A WorldBank report on Tonga stated that, “Large private remittances and officialgrant inflows have enhanced consumption and resulted in negative grossdomestic savings equivalent to nearly a quarter of gross domestic product”(World Bank 1993). Prasad similarly argued that, “The strategy (or non-strategy) of depending on remittances provides crucial revenue and contributesto the balance of payments in small countries, but creates a consumptionsociety, where productive economic activities hardly exist” (2003). Thereare, however, alternative perceptions. It is not easy to assess the extent towhich higher household incomes as a result of remittances affect patterns ofconsumption and savings at the margin, and only one study has sought toaddress this issue in any detail (Walker and Brown 1995). The use ofremittances is also a function of volume and a variety of other factors. Theremainder of this section therefore discusses migrants’ expressed or perceiveduses for their remittances rather than analyzing actual uses.

DEBT REPAYMENT

It is not surprising that remittances are primarily used for consumptiongiven the range of barriers to using them for investment (Ahlburg 1991;Connell 1980), yet in reality, they have different and distinct uses. Migrationcan be costly hence remittances are often first used to repay debts that migrants(and their families) incurred in moving (Tongamoa 1987). This use emphasizessocial obligations. In this sense at least the first wave of remittances constitutesinformal loan arrangements within families (Poirine 1997; Brown and Poirine2004) and is far from being an incidental or random by product of migration.

In parts of Papua New Guinea, migrants even repay their parents for themoney they spent educating them (Morauta 1985). Elsewhere in Papua NewGuinea, Boyd has argued that, “Parents and older siblings stake their claimson the responsibility they shouldered in rearing a future migrant to adulthood(1990). Even more precisely,

The Use and Impact ofRemittances6

Pacific-Remittances.pmd 15/03/2005, 10:56 AM30

The Use and Impact of Remittances

31

Ponam [island] is not a society passively allowing or sufferingthe migration of its members. Rather it is a society which activelyregulates this migration, and it does so in a way which helps securea substantial remittance income [hence people say] ‘Children areour garden and we survive by eating the fruit’ (Carrier 1984).

On Ponam and on other islands like Nanumea in Tuvalu, control of bothmigration and population growth thus has a conscious economic rationalityequating an economic return (remittances) with an economic investment (childrearing for migration). This is more appropriately labor export rather thanmerely migration.

CONSUMPTION

Food

Most studies suggest that food is one of the main items purchased withremittances. Expenditure on food, much of which is imported, has in factbecome so extensive that the cost of food imports is considerable in someisland states. Nowhere is this more evident than on the small island of Ware,Papua New Guinea where households spent fully 88% of all remittanceincome on food (Hayes 1993). A similar situation was true on the outlyingFijian island of Rotuma (Rensel 1993). In Tuvalu, two thirds of respondentsclaimed that food was the most important use and about half of all remittanceincome was spent on it with additional expenditures for electricity for therefrigerators that were consequently necessary (Clark 2004). Much the sameis true in Kiribati where almost all remittances are used for basic needs,including food and clothes, and what is left is saved for almost obligatorycommunity feasts (Borovnik 2004). In both Kiribati and Tuvalu, expenditureon processed food, especially tinned fish and rice, increases immediatelyafter the arrival of remittances (Dennis 2003). In each of these cases, thestrong focus on basic needs, especially food, is due in part to poverty onouter islands and in part to the exceptionally limited alternatives for usingremittances. National estimates for Tonga suggested that about 70% ofremittances was spent on imported tinned and preserved foods, beverages,and tobacco (Tongamoa 1987).

Goods

There has been a transition from using remittances to purchase basicssuch as food, to spending them on items such as outboard motors and

Pacific-Remittances.pmd 15/03/2005, 10:56 AM31

Remittances in the Pacific

32

construction materials. More recent studies have demonstrated not only thatremittances are more likely to be used for modern goods such as sound systemsrather than for basic needs, but that as in the case of Fijians in the UnitedStates these goods include telephones, televisions, and video systems. Indeedin this case and probably others, migrants have been instrumental inconnecting their families in Fiji to themselves and the wider world by financingthe installation of telephone services (Scott 2003). Despite the massive growthof the Internet and e-mail in the Pacific islands, there is no evidence yet thatremittances have been used to purchase computer systems, but this too maybe due to the lack of recent studies.

HOUSING

The construction or purchase of improved housing is sometimes perceivedas a form of consumption rather than as an investment or a welfare gain.Overall, expenditure on improved housing contributes to welfare, security,prestige, and income generation. Moreover, to a greater extent than otherforms of expenditure, it has a multiplier effect as it provides employmentand expands other local industries and services. Using remittances for housingis so widespread in some countries such as Tonga that casual inspection of avillage shows which households have migrants overseas. In the outer islandof Falahola, the largest and most regular remittances were spent onconstruction materials (James 1991). In Samoa, housing improvements alongwith small-scale luxury consumption and church donations represent aninvestment in security (Shankman 1976). In several places including the CookIslands (Loomis 1990), houses built with remittances are rented out andbecome real investments.

SAVINGS

To date there has been very little evidence of remittances directlycontributing to savings or to other financial investments (Ahlburg 1991; Brownand Connell 1993). However, in the case of Tuvaluan seafarers where actualuse is more transparent and where return migration is necessary, as much as37% of remittances went into personal bank accounts rather than to partnersor families (Dennis 2003). In Fusi, Samoa, villagers argued that the amountremitted was too small to meet anything other than its immediate purposeand that any surplus was absorbed by life crises (Muliaina 2001). In Falahola,Tonga, however, “…in remitting to their wives, who bank most of the fundsfor the family’s future, the young husbands are sending money to themselves”(James 1991). Kiribati seafarers may also hold remittances for investment

Pacific-Remittances.pmd 15/03/2005, 10:56 AM32

The Use and Impact of Remittances

33

instead of spending them immediately (Borovnik 2003). These kinds ofsituations are probably more common than is otherwise recorded, though ingeneral, savings rates are low in the Pacific islands.

Households in Tonga that received remittances saved more than thosethat did not, but as would be expected, the inclination to save remittanceswas no different than the inclination to save total income since remittancesare just one source of income among several (Walker and Brown 1995).There is also growing evidence that money that might have been remitted issaved in the host countries for use at a later date. Moreover, for at least Samoaand Tonga, remittances and savings are both linked to interest rates (Foster1995). Thus, despite the focus that much of the existing literature gives to thesocial context and impact of remittances, there are clear indications that bothsenders and recipients behave in an economically rational manner.

AIR FARES AND EDUCATION

In a different sense, remittances are invested in the future in two ways.On one hand, they are used to purchase airline tickets for other familymembers. Some two thirds of Tongan remitters in Auckland claimed to havepaid for at least one airfare for a relative, and some migrants had paid for upto ten. Similarly, 75% of Samoan and 33% of Tongan households had hadairfares paid by a migrant family member during 1992 (Brown 1995). Therecipients of airfares were broadly similar to the recipients of remittances(Fuka 1985; Morton Lee 2004a).

Remittances for school fees are also important and tend to have a standardvalue. Migrants without younger siblings in school do not usually send any.After food, education was the most important use of remittances in Tuvaluand probably in Kiribati since education is specifically regarded as aninvestment in the future (Clark 2004; Borovnik 2003). In these ways and inothers, remittances support the migration-remittance nexus by ensuring thatother family members have the opportunity to benefit from education andmigration.

INVESTMENT

Remittances are used for various forms of investment sometimes inagriculture but more frequently in the service sector especially for shops andtransport businesses. Nevertheless, there is still a general belief that, as inSamoa, “…remittances were not used for capital investment or, to be morespecific, for speculative capital investment” (Shankman 1976). Shankmanwent on to observe that, “…migration was a far more lucrative investment

Pacific-Remittances.pmd 15/03/2005, 10:56 AM33

Remittances in the Pacific

34

than anything available in the village” (1976), hence the conservative use ofremittances reflected limited investment opportunities especially in productivesectors. Since that early study there is, however, evidence there and elsewherethat remittances have constituted the start-up money for many shopkeepersand other small entrepreneurs. Half of all market vendors in Apia, Samoa, allof whom received remittances, claimed that some had been used as capitalfor the purchase of seeds, fertilizer, and tools to produce food to sell (Muliaina2001). Walker and Brown (1995) found that a significant proportion ofremittances received by Samoan and Tongan households was used for businessand farm investments. Apart from this, there is limited evidence that anyinvestment of remittance money has had a lasting impact on future incomegeneration and employment, but this may be because many of the studieswere done in relatively remote locations.

In most places, attempts have been made to invest remittances. Even ona small outer island like Falahola (Tonga), remittances were used for economicventures ranging from agriculture to tourism, though remoteness limited theirsuccess (James 1991). Here at least and no doubt elsewhere, remittanceswere, “...highly individual and many are concerned with capital accumulation”(James 1993b). Throughout Tonga, the increased use of remittances forinvestment in fishing, agriculture, shops, and transport businesses attests tothe shift from consumption to investment (Faeamani 1995; Walker and Brown1995) that has occurred in part as consumption goals have been satisfied.This transition has also occurred in parts of rural Papua New Guinea (Boyd1990) and Pakistan (Helweg 1983) and in several parts of the easternCaribbean in similar small island environments (Connell and Conway 2000).While the transition may benefit economic development, at least oneanthropologist (Small 1997) has raised concerns that it might also furtheremphasize intra-village (and country) economic inequalities and hamper socialdevelopment.

Where there are opportunities and where consumption goals have beensatisfied, remittances are used for investment, to stimulate entrepreneurialand trading activity, to increase formal sector employment, and to producemultiplier effects. The expansion of the Nuku’alofa flea market in Tongasince the mid-1980s is an example. Many households used remittances inkind as investments in their market stalls, and these stalls sometimes enabledeconomic diversification. (Significantly, the flea market was forced to movethree times within a few years partly because of opposition from the privatesector which felt that a “black economy” of this kind might undermine theformal sector). This indicates a shift away from using remittances forimmediate consumption toward increasing their returns throughentrepreneurial activity (Brown and Connell 1993). Moreover, James noted

Pacific-Remittances.pmd 15/03/2005, 10:56 AM34

The Use and Impact of Remittances

35

from a parallel perspective that, “It has been argued that remittances takeaway the motivation of locals to produce, but the facts of localentrepreneurship seem to contradict this since large consignments of localproducts. . . are sent to relatives overseas for sale among the Polynesianpopulation” (1991).

In certain contexts there is also an urban bias in the use of remittances.In both Kiribati and Tuvalu, though overseas jobs are rotated among islandsso that residents of outer islands have access, one outcome is that theimmediate families of some of the migrants move into the capital. Moreover,when seafarers return, a substantial proportion of their income is spent eitherin South Tarawa or Funafuti rather than on the outer islands. In Kiribati atleast the majority is spent in South Tarawa (Clark 2004; Borovnik 2003).Indeed in certain circumstances relatives migrate into town with the specificobjective of laying claim to some of the overseas earnings.

COMMUNITY USE

A proportion of remittances is invariably used for community objectivessuch as support for social organizations and institutions, usually churches.Both in Samoa and Tonga (Shankman 1976), substantial sums have beenremitted directly or indirectly for church construction and maintenance,pastors’ incomes, and related activities. This also appears to be the mostimportant goal for remittances sent by Fijians in the United States (Scott2003). Brown (1995) found that 41 and 18% of Samoan and Tongan migrants’remittances, respectively, were sent to nonhousehold institutions includingthe high schools and primary schools run by the churches (Tongamoa 1987;Faeamani 1995; James 1997; Sudo, 1997). On the Tongan island of Ha’ano,the most significant correlation between remittances and other factors waswith church expenditure (Evans 2001). In Tonga, most gifts to the church aresent to parents or close relatives to be given in the annual, highly competitive“free-gifting” ceremony, the misinale. The amount of the donation isannounced in church and the whole family derives prestige (or, alternatively,notoriety and shame) from the extremely visible public presentation.

Remittances are also directed to local sporting associations, women’sgroups, specific local development projects, scholarships and so on, whichin many cases point to wider social gains. In a sense therefore, Tongansoverseas support Tonga by sending money for various, usually village projectslike schools or water systems and public occasions. “Overseas Tongans mightrepair the local island church, or contribute to the centennial celebration oftheir high school, or pay for the school fees of a relative who has passed herexams for the next grade level” (Small 1997). Samoans have similarly

Pacific-Remittances.pmd 15/03/2005, 10:56 AM35

Remittances in the Pacific

36

contributed to a range of village facilities including schools, dispensaries,fish traps, access roads, and electricity generating plants (Macpherson 2004).However, in Tonga at least, sending money through the family for publicpresentation emphasized that part of the rationale was to give the familystatus in the community as pious, generous, prosperous and “properly Tongan”and only secondarily to support the social group (James 1997). In Tuvalu,remittance recipients are expected to support the relevant community islandfund, and women contributed to this as a matter of pride even when theylived on the capital island of Funafuti. There were, however, householdbenefits to this since it demonstrated that the seafarer acknowledged the peoplewho were taking care of his land until he retired from the sea (Clark 2004).

SOCIAL USES

Remittances are used for a variety of social purposes including meetingthe high costs of weddings, funerals, and other ceremonies but also for thecosts of education that may lead to long-term economic gains. In mostcountries, contributions to significant life-cycle events are essential formigrants to retain a basic stake in home society, and requests are very oftenspecifically for such functions. Samoan migrants have also contributed tofamilies’ and parents’ attempts to enhance their status by supporting theirbids to obtain chiefly titles and parliamentary seats (So’o 1998). It isparticularly through these kinds of social expenditures that migrantsemphasize their continued stake in and commitment to the home society.

Unequivocally, investment in flexible human capital is a most importantand highly valued use of remittances, especially in circumstances where thelack of skilled individuals is a constraint to development. Indeed, much ofthe literature on migration has neglected the massive social significance ofremittances by emphasizing their more obvious economic effects. Walkerand Brown (1995) stated that for Samoa and Tonga, social uses predominatedin 63 and 79% of households, respectively. Overall, nearly 10% was used forsuch purposes.

In places where remittances constitute an important part of total householdincome, it can be expected that they will also have a significant impact on itsdistribution though there is no consensus on how remittances actually affectdistribution. This is not altogether surprising given the variety of forms,channels, and uses of remittances, both formal and informal, and the lack ofgood data on household income and expenditure. However, until relativelyrecently, the dominant view was that remittances were likely to reinforceincome inequality by enhancing the capacity of recipient households to investin additional migration, education, and other income-generating assets

Pacific-Remittances.pmd 15/03/2005, 10:56 AM36

The Use and Impact of Remittances

37

(Connell 1980; Shankman, 1976). Some village studies have demonstratedconsiderable income inequality (Hardaker et al. 1987) and have suggestedthat this is partly a result of remittance flows (Gailey 1992a; Small 1997).However, more recent empirical studies based on survey data have tended tochallenge this view. From independent data sources, Ahlburg (1991, 1995)and Brown (1995) found that distribution of household income withremittances was less skewed than the distribution without remittances.Furthermore, other recent studies have indicated that inequality is a functionof many factors of which the migration-remittance nexus may be anunimportant or tiny part (Evans 2001; Muliaina 2001; Halatuituia 2001).

Pacific-Remittances.pmd 15/03/2005, 10:56 AM37

Remittances in the Pacific

38

Despite evidence of the growing significance of investment, it has oftenbeen argued that remittances (and aid) are not conducive to some forms ofprivate sector development and thus keep the economy in a state of limitedproductivity. This is broadly what Ahlburg (1995) has called the “disincentiveeffect.” Indeed the MIRAB model implies that a strong bureaucracy maystifle rather than increase productivity. This perspective was stressed by aseries of authors in the 1990s (MacMaster 1993; Browne 1995; Duncan 1994;Pollard 1995; La Plagne et al 2001). They also drew parallels to the mannerin which a booming sector like mining in Papua New Guinea produced theDutch disease by driving up wage rates (Auty 1993). In the case of Tonga,Sturton argued that,” The Tongan economy displays all the characteristicmarkings of the ‘Dutch disease’ where a dominant export activity attracts adisproportionate command over resources, pushes up domestic productioncosts, and reduces international competitiveness. In the Tongan case the‘booming’ sector has become development assistance and migrants’remittances” (1992). Similarly Faeamani argued that through the combinationof the loss of young adults and an inflow of cash in the form of remittancesand goods, “…there is a consequent reduction in garden size and production”(1995, see Fairbairn 1993). More generally, several authors have stressed thewide-ranging notions of dependency that remittances appear to create.

Several authors have specifically stressed that the crowding-out effect ofremittances combined with notions of an easy subsistence lifestyle oncecharacterised as “subsistence affluence” (Fisk 1995) together discourageproductivity. Thus MacMaster suggested that in the Cook Islands, Samoa,and Tonga, remittances are, “…a mixed blessing as they undermine theincentive to work and are rarely spent on productive investment. They arenormally used for unproductive ceremonial purposes or on imported luxuryconsumption items” (1993). The World Bank argued that, “…remittanceinflows have had an adverse effect upon labor supply leading to higherreservation wages and a corresponding reduction in the production oftraditional export crops” (1990). These and other similar statements andconclusions (e.g. Ahlburg 1991, Finau 1994) have overall suggested that,“...it is not clear that the net effect of remittances and aid is conducive to longterm economic viability and prosperity” (Cuthbertson and Cole 1995) if they

The Disincentives ofRemittances?7

Pacific-Remittances.pmd 15/03/2005, 10:56 AM38

The Disincentives of Remittances?

39

reduce the benefits from investments and drive out private capital. Few ofthese studies, however, present data that justify these assumptions andconclusions.

A number of the more extreme critics of aid and remittances may beplaced among the subsistence affluence school of Pacific paradise dreamingwhere, as Scott has argued, “It is all too easy and a serious mistake toromanticise the social arrangements that distinguish much of peasant society”(1976; Connell 2003b). It simply cannot be assumed that there is no desire tomaximize or even improve incomes and that islanders are somewhat irrationalor even lazy. In fact, subsistence activity is only exceptionally abandoned orreduced despite receipt of remittances. More significantly, where there areundeniable possibilities for production, participation is considerable. Thedevelopment of new cash crops such as coffee in the Papua New Guineahighlands, kava in Vanuatu and Fiji Islands, and vanilla and squash in Tongasparked enormous spontaneous interest and the rapid growth of theseagricultural industries. The rise of squash in Tonga occurred during timeswhen remittance levels were high. The principle participants in the industryhad a secondary or tertiary education, overseas experience, and a history ofemployment with the government, but also had some traditional attitudes.Moreover the growth of the industry was, at least initially, stimulated bygovernment involvement (van der Grijp 1997, 1999). An inverse relationshipbetween squash export earnings and the level of remittances followed (Overtonet al 1999) again emphasizing the particular economic significance ofremittances and also the manner in which remitters respond to needs. Thuswhere the conditions are appropriate, even if remittances have reached highlevels, the private sector may flourish and be stimulated.

The crowding out argument was prominent in the initial formulation ofthe MIRAB model by Bertram and Watters who argued that with additionalsources of income from remittances (and indirectly from aid) people, “...canbe expected to evaluate the return on [agricultural] investments relative tothe alternatives. On this basis it would be expected that as the alternatives tocommercially oriented agriculture would improve, so a reallocation ofhousehold effort away from agriculture would take place” (1985). As notedabove, there is not a great deal of evidence that this is what happens in mostplaces. Faeamani’s (1995) observations of declining production in Tonganvillages may be exceptional and indeed may more likely be a function oflabor shortages (Evans 1996, 2001). In circumstances where both remittances(and wages) are high and there has been substantial migration, middle agedand elderly residents may abandon full-time agriculture and become moredependent on remittances or other sources of income as has long been thesituation in the Cook Islands (Curson 1979) and in Niue (Cohn 2003), but in

Pacific-Remittances.pmd 15/03/2005, 10:56 AM39

Remittances in the Pacific

40

both these countries populations are absolutely declining. Such situations arerare and probably much less common than those where remittances havebeen used for investment in agricultural projects (Connell 1980). Even so, ifthis is what has occurred as populations in the Cook Islands and Niue fall, itis nonetheless exactly, “…what is required for efficient economic behaviour:that the family allocates its resources to the highest productive use, even if ithappens that this particular use is not ‘productive’ in the ‘domestic’ economy,but rather in a ‘foreign’ economy” (Poirine 1998).

In contrast, even those who have criticized remittances as brakes onindependent development and causes of accelerated dependency noted thathigh levels of dependence on remittances may not affect farmers’ responsesto price incentives or their interest in new crops (Shankman 1976). O’Mearareached similar conclusions in Samoa while also observing that levels ofproduction were tied to the rate of return. If levels of return were low andfalling, villagers reasonably sought to diversify their incomes away fromagriculture towards a range of sources including remittances (1990). Morerecently it has been noted that food production has been less enthusiasticallypursued in Samoa but, despite an expected correlation, there was no evidenceto link this with remittances. The principal reasons were labor shortages andtaro blight (Muliaina 2001). Much the same conclusions were true in Tonga(Evans 2001). In both cases, however, labor shortages were linked tomigration.

Most studies of remittances have observed that a significant proportionsupport traditional customs and obligations. As Evans has stressed, this is atleast in some part because economic opportunities are few, so investing incustom avoids what would amount to, “…intensive self-exploitation inagricultural activity” and gives villagers both respect and autonomy (2001).Similarly, sellers in the Nuku’alofa flea market whose goods mostly arrivedas remittances and who might be seen to be involved in trade, investment,and market capitalism preferred instead to see themselves as part of complexreciprocal exchange systems that, “…maintained the social integrity of Tongansociety despite diasporic fragmentation” (Besnier 2004). It is simply moreappropriate to engage in exchange and gift giving rather than sales andpurchases. Commercial practices are therefore downplayed in favor of socialobligations. Alexeyeff stated many local and overseas Cook Islanders,“...considered economic transactions as emotion-driven actions” (2004). Thesocial and the economic cannot be disentangled.

International migration in the quest for remittance incomes is a consciousstrategy for households and even for countries like Kiribati and Tuvalu thatmakes economic sense in small and open economies. Moreover, as almostevery study of remittances has revealed, even in the most difficult

Pacific-Remittances.pmd 15/03/2005, 10:56 AM40

The Disincentives of Remittances?

41

circumstances recipients make efforts to invest where they can. In the outerislands of Kiribati where most remittances provide basic needs and the customof bubuti (a request that cannot be refused) makes savings let alone businessalmost impossible, all recipients nevertheless sought to retain some incometo invest in land, doughnut bakeries, stores, or even in sewing material forblouses that might later be sold (Borovnik 2004). Here as elsewhere, there isno evidence that any part of the economy is abandoned or neglected butrather that remittances enable some form of diversification and investmenteven in the most difficult circumstances.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM41

Remittances in the Pacific

42

The erroneous supposition that remittances are used primarily for theconsumption of imported consumer goods (World Bank 1991, Fairbairn1991a, 1991b) has increased concern among policy makers and internationaldonors that they have little or no impact on domestic investment and thus dolittle to enhance the longer-term economic sustainability of island economies(Forsyth 1992; Cuthbertson and Cole 1995). Though many studiesdemonstrated, sometimes in considerable detail, that consumption is ofconsiderable importance (e.g. Faeamani 1995), they also showed that higherlevels of consumption were a major welfare gain and contributed to meetingbasic needs that were otherwise sometimes poorly satisfied (Tongamoa 1987).They also showed that remittances have been increasingly important for bothinvestment and savings (Walker and Brown 1995) as the entrepreneurialbehavior of Nuku’alofa market vendors attests. Further support for thistransition from consumption to investment comes from Foster’s (1995)econometric analysis of secondary Tongan and Samoan data which suggeststhat remittances are sensitive to interest rates because of the desire of migrantsto hold assets. There is, however, varied support for the view that as migrantincome grows, remittances also grow, and that they will grow much faster ifinterest rates are favorable. The scant evidence that exists does suggest thatusing remittances for investment and for savings increases over time, eventhough in small, isolated island states there are very substantial constraints tosuch productive use. Remittances are also invested in economic activities, inhuman capital, and in the well being of others. Some of this perpetuates themigration remittance nexus and increases the flow of funds.

Other important findings have emerged from studies of migration andremittances in the region. It has become apparent that the official, aggregatedata on remittances, income, and savings provide a highly misleading pictureof the actual extent, form, and use of remittances. Remittances appear to bemaking a substantial contribution toward savings, but this is not reflected inthe official data because of the form in which savings are often held andbecause of the inadequate, inconsistent, and sometimes quite inappropriatemanner in which aggregate savings rates are calculated and treated in mostmacroeconomic analyses. Cross-sectional data from recent studies stronglyindicate that remittance levels do not appear to decline with length of absence

The Policy Context8

Pacific-Remittances.pmd 15/03/2005, 10:56 AM42

The Policy Context

43

from the home country and that an increasingly significant motivating factorfor migrants to remit is the accumulation of assets and investment there.Both of these conclusions are contrary to many previous studies of migrationand remittances in the island Pacific and in many other parts of the world.

Several recent studies have demonstrated that the remittance decayhypothesis has limited validity, at least as far as the remittance behavior ofindividual migrants is concerned. Although Tongamoa (1987) found someevidence of support for this among Tongan migrants in Sydney, other studieshave found that decay was minimal for Cook Islanders, Fijians, Samoans,Tongans, and Tuvaluans in various cities in Australia and New Zealand (Fuka1985; Loomis 1990; Stanwix and Connell 1995; Simati and Gibson 2001).While these were the results of relatively small sample surveys, similar andmore statistically reliable results were derived from a more detailed study ofSamoans and Tongans in Brisbane (Walker and Brown 1995; Brown 1998).These data show few signs of the remittance decay that in other studies mayhave been confused with the effect of slowing income growth amongindividuals and households in host countries. In fact, remittances did notappear to decay despite higher levels of unemployment and lower averagereal wage levels, offering evidence of the sacrifices made by some migrants.As income rises, saving appears to become more important in determiningremittances in both host and home countries.

Consequently, it is neither the availability of savings nor the unwillingnessof recipients to invest that explains the relatively poor performance of smallPacific island economies. The development problem is as much one ofresource use and investment allocation as it is of generating savings. Thissuggests that there are areas where public policy interventions would beappropriate as they have previously been in Asia (Brown 1994, 1995; Russell1986, 1992). Generally in the Pacific, there have been very few attempts tointervene in the, “…normal laissez-faire structure of migration; those thathave been made have been limited in their extent, effectiveness andapplicability” (Connell l987). Even attempts to influence the structure ofinternational migration other than to promote it, let alone to direct and benefitfrom the flow of remittances, have been conspicuous by their absence.

Three key policy contexts exist. The first is accommodating measuresdesigned to remove or minimize problems with migration. In an internationalcontext, these measures have included employment agreements with hostgovernments and some of the Polynesian countries. For some time, Tongahad such a relationship with New Zealand (Connell 1983), but it is now oflimited significance. The labor migration to Nauru from Kiribati and Tuvaluis of greater contemporary importance, though this has virtually ended. Kiribatiand Tuvalu and to a lesser extent the Marshall Islands, Wallis and Futuna

Pacific-Remittances.pmd 15/03/2005, 10:56 AM43

Remittances in the Pacific

44

(and perhaps Samoa and Tonga), are the only countries in the region thathave actively encouraged international labor migration although others, likeCook Islands, have not usually sought to discourage the flow. (Since the1990s, Niue has discouraged migration, encouraged return, and rewardedimmigration with very high per capita aid because the population has fallenbelow 2000 and may not be sustainable. This pressure increased after CycloneHeta in 2004). Pacific island states overall have therefore effectively soughtto achieve accommodation through increased migration. Indeed in situationswhere remittances are rather more stable and sustainable over time thanreceipts from agricultural exports or tourism, as is the recent experience ofCook Islands, Samoa, and Tonga, it is probably an optimal policy to specializein labor export (Poirine 1998).

Only two national development plans, Samoa’s Third NationalDevelopment Plan of 1975 and to a lesser extent Tonga’s Sixth DevelopmentPlan of 1991, have commented on the negative impact of internationalmigration, and hence the need to minimize it, though Tonga’s plan noted thatthe only real negative feature of international migration was the failure tomaximize its benefits.

A large proportion of Tongan migrant remittances is spent on consumergoods and not channelled into investment for the development of theproductive capacity. Savings bonds denominated in foreign currency couldbe issued and offered to non-residents. This development would imply thatreturns on domestic securities be consistent with those offered to Tonganmigrants, as Tongan residents would otherwise be encouraged to seek accessto non-residents’ instruments (Tonga 1991).

There is no evidence that any of the proposed policy interventions wereseriously considered, let alone implemented despite their perceived urgency.Concerns over the loss of skills were also stressed in the 1980 and 1992Samoa Development Plans but have not been subsequently repeated.Elsewhere in the region they have rarely if ever been discussed because ofthe economic benefits from migration. Generally, in fact, there have beenfew attempts to direct or influence the structure of international migration inthe South Pacific due to official respect for the rights of individuals to movefreely.

Secondly, there is a range of development policies that might increasethe gains from international migration that could be initiated by home countrieswith the cooperation of host countries. Schemes where groups of labormigrants from a number of countries travel overseas for several months togenerate funds from donations or employment for specific local social oreconomic development projects might make better use of remittances. Thispossibility is under consideration in Australia (Australia 2003).

Pacific-Remittances.pmd 15/03/2005, 10:56 AM44

The Policy Context

45

Thirdly, there are policies that might influence the productive use ofremittances, yet it is also clear that existing financial systems and relatedpolicies do not attract remittances into financial assets in the home economies.Real interest rates have been negative or very low. This has encouragedrecipient families to invest in noninterest bearing assets such as housing orconsumer durables. Migrants may, quite rationally, choose to hold theirsavings abroad, perhaps to be remitted home and invested at some point inthe future.

POLICIES TO FOSTER REMITTING THROUGHOFFICIAL CHANNELS

It is generally accepted, at least informally, that policies are needed toencourage the use of remittances to promote longer-term growth and incomesecurity in home economies. Pacific island countries have yet to developpolicies that (i) send more remittances through official rather than informalchannels; (ii) increase the levels of remittances by encouraging migrants tohold their savings in financial assets in the home country rather than keepingthem abroad (or spending their savings on consumer goods); or (iii) encouragemigrants to become investors in productive assets in the home economies.

Governments of labor-exporting countries elsewhere in the world haveintroduced a variety of schemes with these policy objectives in mind, namely(i) repatriable foreign exchange accounts to encourage the greater use ofofficial channels, (ii) foreign currency denominated bonds to encourage moreuse of financial assets in the home country, and (iii) self-employmentinvestment schemes to stimulate more direct investment in productive assets.Athukorala (1993) reviewed policies in seven major labor-exporting countriesin Asia. All except Indonesia provided temporary and permanent migrantworkers with the incentive to remit to repatriable foreign currency accountsin domestic banks which effectively means that the migrant is not subject toforeign exchange controls in current account transactions and capital transfers.In addition, India and Pakistan offered premiums over the interest ratesavailable in the international financial market. Bangladesh offered additionalincentives through a preferential exchange rate on conversions of foreignexchange from the repatriable foreign currency account to local currencyand a wage earners’ scheme that enabled migrants to sell their foreignexchange to importers at daily auctions (Mahmud 1989). In Pakistan, therewas the added advantage that the Habib Bank was allowed to open branchesin many of the labor importing Arabian Gulf countries to facilitate migrants’use of official channels (Abella 1989). Sri Lanka also offered its migrantsduty free domestic shopping (Saith 1989). Since most migrants from Pacific

Pacific-Remittances.pmd 15/03/2005, 10:56 AM45

Remittances in the Pacific

46

island states are settlers rather than labor migrants some of these provisionsare not relevant.

In other instances, governments have resorted to mandatory remittanceratios. For instance, the Philippines introduced a decree in 1983 that requiredmigrants to remit a given percentage of their foreign earnings through officialchannels to be converted to domestic currency at the official exchange rate.This proved both unpopular and impossible to implement and was scrappedin 1986. A mandatory remittance scheme was also adopted by Thailand butwith little success (Quibria 1986, Quibria and Thant 1988).

In order to encourage migrants to hold their savings in financial assets intheir home rather than host countries, many governments have introducedforeign currency denominated bonds. Athukorala (1993) pointed out that thesehave an added advantage over RFCAs in that they guarantee anonymity ofthe asset holder who is still entitled to repatriate the funds when the bond isredeemed. For instance, Bangladesh has a wage earner development bondoffering interest rates above those on domestic bonds as well as an insurancescheme based on a one-off premium in foreign currency. Pakistan has forsome time had Khaas deposit certificates denominated in local currency andissued to migrants on payment of foreign exchange that upon redemptioncan be converted back to foreign currency at the official exchange rate (Kazi1989). Somewhat more recently, Pakistan introduced foreign exchange bearercertificates that carry an interest rate above the Euro rate (Kazi 1989), and adollar bearer certificate offering a return linked to the London interbank offeredrate (Kardar 1992; Athukorala 1993). India has a similar scheme (Gordonand Gupta 2004).

The third policy area concerns schemes to encourage migrants themselvesto become investors. The first country to introduce such measures was Turkeythrough the formation of village development cooperatives whose membersthen gained preference for migration (Swamy 1981, Russell 1986). Pakistanhas a scheme that allows migrants to import machinery at concessional ratesof duty and to invest in export processing zones (Kazi 1989). In Bangladeshtoo, migrants have been offered special fiscal incentives to invest domestically(World Bank 1981).

Despite very different socioeconomic conditions in the labor-exportingAsian countries, their experience cannot be ignored. The findings of the mostrecent studies in the Pacific indicate that there is substantial scope forgovernment policy interventions to increase flows of remittances to theireconomies, yet there has been virtually no concerted effort by any governmentto offer incentives for migrants to remit more through official channels. Allthe evidence suggests that migrants’ remittances would be responsive tofinancial incentives of the sort that have been adopted in Asia (Foster 1995).

Pacific-Remittances.pmd 15/03/2005, 10:56 AM46

The Policy Context

47

While it needs to be acknowledged that this runs counter to the conclusionsfrom econometric analysis of secondary data in other remittance receivingcountries (Swamy 1981, Straubhaar 1986), it also needs to be stressed thatthese studies relied exclusively on secondary macroeconomic data.

Policy makers must therefore regard remittance levels as potentiallyresponsive to policy interventions and the provision of special incentives.Until such time as the financial authorities in the South Pacific offerinternationally competitive real interest rates to savers, migrants may choose,quite rationally, to hold their savings elsewhere. Indeed there is some evidencethat although extended families are behaving to some extent as transnationalinvestment institutions (Marcus 1981), the countries of the Pacific areunwilling to develop policies that may be seen to favor those who have, throughinternational migration, achieved a greater degree of success than those whoeither chose to remain at home or were unable to migrate.

There is scope for a careful consideration of policies, such as the transferof pension rights, to maximize the benefits of international migration in mostcountries in the region. This is true even though migrants from Pacific statesare more likely to be permanent residents. One strategy that might be ofvalue is attracting and retaining stocks of migrant savings for investment. Inhost country recessions, it is likely that remittances for saving and investmentwill be reduced instead of those for family consumption. Channelling migrantsavings to home country assets can provide security in such situations.

POLICIES TO FOSTER SAVINGS AND INVESTMENT

It is generally recognized that policies to promote remittances and to channelthem into more productive investments have not met with tremendous success.One of the most cited studies in this regard is Swamy’s (1981) study for theWorld Bank. Using pooled time series data on official remittances from threecountries (Greece, Turkey, and Yugoslavia) over 18 years, she found that policymeasures such as relative interest rate schemes and premium exchange rateswere unsuccessful in increasing remittance flows. Another study using secondarytime series data from Turkey showed that, “…neither variations in exchangerates (reflecting government intention to attract remittances by premiumexchange rates) nor changes in the real return on investments (reflectinggovernment intention to attract remittances by foreign exchange deposits withhigher returns) turned out to affect the flows of remittances (Straubhaar 1986).Investment schemes like the village cooperatives attempted in Turkey also hadlittle success (Swamy 1981; Russell 1986). In Asia too, Saith (1989) pointed tothe high failure rate of the self-employment schemes attempted by governmentsof labor-sending countries to convert return migrants into small entrepreneurs.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM47

Remittances in the Pacific

48

In the Pacific, there have been virtually no concerted efforts bygovernments to offer incentives for migrants themselves to invest remittancesin productive activities. It has not therefore been possible to assess policyinterventions from an ex post perspective. The limited success of policiesadopted in Asia suggests that it would not make good sense to replicate them,but on the other hand, it is not necessary to dismiss them. Much can be gainednot only in the Pacific from analysis and assessment of the potential of thepolicy environment. First, the conditions under which policies would beeffective must be identified, and second, the extent to which such conditionsexist must be ascertained.

It is useful to begin by identifying the assumptions underlying policiesto stimulate a greater flow of remittances into sustainable investments thathave been adopted elsewhere.

(i) Investment is limited by savings and/or the availability of foreignexchange. (If this were not the case there would be no reason tobelieve that increasing the inflow of migrants’ remittances wouldinduce, or at least, enable, additional investment).

(ii) Investment in the economy of the home country is necessary if thelong-term income security of the population is to be sustained.

(iii) Migrants themselves are the appropriate agents for investing theirremittances; they are all, in effect, latent entrepreneurs.

(iv) If remittances are to be channelled into productive investments inthe home economy, they must be transferred through official channels.

(v) Migrants’ savings in the home country and their remittance levelsare sensitive to relative real interest rates. This, in turn, implies thatmigrants are motivated to remit for reasons of self interest (financialgain) and not only to meet the needs of nuclear and extended families.

The most recent economic studies provide considerable information onthese assumptions. First, it is now evident that migrants’ remittances arepotentially responsive to financial incentives since they remit not only forfamily support and that a significant proportion of their remittances appearsto be motivated mainly by investment. Second, the fact that a significant partof unofficial remittances goes into investment goods suggests that remittanceshave a greater impact on domestic investment than is generally thought andthat they need not be transferred through formal banking channels for thispurpose. Third, from surveys of both migrants and recipients, it is evidentthat some are saver rentiers, others are saver investors, and some, like theTongan flea market vendors have evolved into full scale entrepreneurs withinvestment links with other sectors of the economy (Brown and Connell 1993).

Pacific-Remittances.pmd 15/03/2005, 10:56 AM48

The Policy Context

49

From the micro survey data available, it is not possible to assess theextent to which suitable avenues and opportunities for investment in thedomestic economies of Pacific island states exist and the extent to whichthey are limited by savings and foreign exchange. Foster’s (1995) econometricanalysis of secondary data on savings and remittances in Samoa and Tongaquestioned assumptions that investment was limited by saving levels, thatthe most appropriate use of migrants’ savings was domestic investment, andthat migrants would necessarily make the best entrepreneurs. In practice,real interest rates in Samoa and Tonga (and probably elsewhere) have beennegative or very low which could account for the shortage of loan funds.These findings led Foster (1995) to conclude that attractive interest rates arelikely to attract greater savings.

It is therefore reasonable to conclude that it is neither the availability ofsavings nor an unwillingness to invest on the part of recipients of privateremittances that explains the relatively low growth performance of theSamoan, Tongan, and other home economies. Where this is the case, it appearsthat the problem is more one of resource use and investment allocation thanof savings. In this context, the role of financial institutions becomes veryrelevant. One possibility that warrants further investigation is the scope forlinks between remittances and microfinance institutions as an alternative tostate and commercial banks.

Financial institutions could play a more active role in mobilizinginvestment funds by providing stronger incentives to individual savers throughmore attractive real interest rates. However, family security and developmentpriorities should, wherever possible, be kept separate. Economic developmentprojects should, as now, be funded mainly through foreign aid and kept quitedistinct from commercial lending for normal purposes. It is inappropriate tosubstitute migrant funds for aid flows in development.

None of the recent studies concludes that remittances should be directedtowards domestic investment and that the migrants (or their families) wouldnecessarily make the best entrepreneurs though some have indeed becomeentrepreneurs. Where opportunities arise, remittances can be motivated byand used for investment (Faeamani 1995; Walker and Brown 1995; Brownand Connell 1993). Yet even where migrants possess the necessaryentrepreneurial potential, if the general investment climate in the homeeconomy is not conducive it cannot normally be expected that migrants willrisk their savings when safer alternatives exist elsewhere.

In several Asian countries, Saith (1989) questioned the wisdom ofadopting policies to convert migrant-savers into migrant-investors. It wouldinstead make better sense for policy to be geared more towards the majorityof migrants to encourage them to become more active in domestic capital

Pacific-Remittances.pmd 15/03/2005, 10:56 AM49

Remittances in the Pacific

50

markets as saver rentiers. As suggested previously, this would necessitatePacific island economies offering savers competitive real interest rates. Theloan funds accumulated in this way could then be invested either in largerdomestic projects, or, where no suitable opportunities exist, they could beheld as overseas assets denominated in foreign currency at the best possiblerate of return. The accumulation and disbursement of loan funds into viableinvestment projects has been hampered by past government developmentand monetary policies. Fostering investment requires the formation of anindependent financial institution that uses best practices to attract funds frommigrants and to lend them domestically and internationally.

Direct financial incentives might be provided for investing remittancessuch as national development bonds from national development banks. Thishas never been seriously contemplated in the region although there wereattempts in Samoa at the start of the 1980s to encourage skilled migrants toinvest in housing and thus increase their commitment to the country. TheBank of Tonga has designed saving instruments specifically for migrantsthat include an international interest differential (the Bank of Tonga targetsavers’ rate minus the Australian fixed deposit rate). In 1991, it also offereda “return to Tonga” savings account, but was on the point of discontinuingthe scheme in 1995 because it had not attracted much capital (P700, 000)possibly because it was unable to maintain the confidentiality of the accounts.These policies raise questions of equity and preferential treatment for returnmigrants or overseas residents, and Pacific island states have hitherto notchosen to move in that direction. Moreover, some countries may even wishto oppose special provisions for migrants in circumstances where limiteddevelopment and investment opportunities exist.

The accumulation of financial assets abroad could provide income securityfor migrant households and a buffer against adverse developments likecyclones in the home country and recessions and rising unemployment in thehost country. Indeed, Kiribati and Tuvalu have such buffers in place, and thenorthern Micronesian states are moving towards similar schemes. Kiribatiset up a Revenue Equalisation Reserve Fund in 1956 as a trust fund to bebuilt up from the profits from its nonrenewable phosphate exports. Tuvaluestablished the Tuvalu Trust Fund along similar lines in 1987. These fundsare invested abroad and have achieved high rates of return that in Tuvalu’scase averaged 19% in the first 4 years of operation. The revenues from thefunds are used for national budgets. Nevertheless, though both countries havegrowing trust fund balances to provide national economic security, both havesignificant socioeconomic difficulties and no clear strategy for using theirconsiderable assets to promote socioeconomic well being without riskingreductions in much-needed aid.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM50

The Policy Context

51

MONETARY AND ANTI-INFLATION POLICIES

The governments of Pacific island states need to implement monetarypolicies that result in positive, internationally competitive interest rates.Without positive real interest rates, individuals and families cannot be expectedto hold savings except in the short term. Clear differences in the inclinationto save were evident in both Samoa and Tonga in line with differences in realinterest rates (Foster 1995).

Rules are required for granting subsidies and guarantees to keep alllending rates low, not just rates provided by aid supported developmentbanks that already have interest subsidies. Legislation would have to bedrafted so that all financial institutions could hold overseas assetsdenominated in foreign or local currency. It would be best if relevant financialinstitutions were run as independent state corporations along the lines ofexisting national provident funds despite their somewhat checkered history.It is clear from the experience of Kiribati and Tuvalu (if not Nauru) thatgovernments can handle overseas investments very effectively in conjunctionwith a reliable agent. Governments should not be able to borrow funds directlyfrom these financial institutions in order to protect investors and to preventcrowding out of private investment.

Development and national security considerations often necessitate budgetdeficits. In the case of small Pacific island countries, aid can cover suchdeficits, but there is danger of inflation since aid increases liquidity. Thismay worsen the trade deficit or drive up prices, particularly when there is noimmediate increase in the supply of goods and services from developmentspending. This undermines savings flows since it drives real interest rates tolow or negative levels. Funds channelled through development banks maydissipate into consumption subsidies and reduce the propensity to savedomestically. In such circumstances, aid should be used mainly for labor-intensive infrastructure projects to generate employment and for fundingstate pensions for the old and education for the young. Funding businessinvestment should be strictly the province of financial institutions that recycledomestic and migrant savings. Governments and aid agencies are notcompetent to do this.

There is little doubt that attractive interest rates induce savings, eitherdirectly or indirectly, and that this is advantageous. If holding overseas assetsis permitted, central banks need not be concerned with liquidity overhang.There is no simple link between the volume of financial assets and inflation.The inflation experience of Pacific island states suggests that it has beenrelated to aid-funded government spending, to rises in import prices, and tonatural disasters, not to large budget deficits.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM51

Remittances in the Pacific

52

It is appropriate that central banks should continue to avoid persistentbudget deficits but should also promote savings as insurance for the future.The evidence presented here offers some indication that as their incomesrise, migrants accumulate financial assets provided interest differentialswarrant it. Evidence from Walker and Brown’s study (1995) suggests thatmigrants hold significant quantities of savings in their countries of residence,so the potential for accumulating savings in the home country clearly exists.The extent to which migrants might repatriate savings in different financialcircumstances is obviously unknown, and there are grounds for caution overthe degree to which migrants might welcome certain kinds of investment anddevelopment in their absence.

Recent studies also point to the need for financial policy reforms thattake into consideration the real rate of return for the saver in relation to ratesof return available to the transnational family in both financial and non-financial assets, both domestically and abroad. The details of these need tobe worked out on a country-by-country basis, taking into account the particularfinancial, institutional, and policy environment of each case and the forms,channels, and determinants of remittance flows.

Conflict can arise between savings and investment strategies and anti-inflation measures. However, during the 1990s, inflation subsided as a centralproblem in many countries providing an opportunity to reorient economicpolicy towards growth and sustainability. The proposals suggested hereconcerning saving and investment do not pose any risk of inflation. Inflationis not caused by the public holding stocks of financial assets provided theyare not lent to government to finance expenditure and provided that suchassets are not used as a conduit for lending aid funds with a low probabilityof repayment (Foster 1995). Aid funds should continue to be managed bydevelopment banks or agencies.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM52

The Disincentives of Remittances?

53

This report has provided an overview of migration and remittances inthe Pacific with particular reference to Samoa and Tonga, two of the largerPolynesian states where international migration and remittances have beenof the greatest importance and where this has been documented to a moresubstantial and credible extent than elsewhere. A number of conclusions aresignificant.

• There has been a consistently substantial and growing volume ofremittances especially in the Polynesian states making up a significantpart of national income in excess of the value of exports and aid.

• The use of these remittances has gone through a partial transition fromconsumption to investment as many consumption goals have been metat least in part. Such trends have occurred elsewhere (Helweg 1983;Connell and Conway 2000) and have boosted standards of living.

• Remittances have been particularly important in the most remote islandswhere development needs are less well met. Even in the most remotelocations such as Falahola in Tonga and outer Kiribati islands, peoplehave invested remittances though the opportunities are very few (James1993b; Borovnik 2003).

• Remittances contribute to valuable objectives such as human resourcedevelopment and are a means of maintaining social networks andcreating social capital (Grieco 2004). In several places, especiallysmaller islands, education is highly valued both in a general sense andfor the development of specific skills in order to create human capitalfor potential migration.

• Overall various studies suggest that remittances are positive andsatisfying for households but insufficient in and of themselves toinfluence national development goals. Economic growth has been verylimited in Pacific island states as Bertram has pointed out, “…becauseof low capital absorption capacity, due partly to small scale andgeographical isolation, which limits the possibilities for text-bookgrowth models based on large country experience” (Bertram 1999b:338; Tisdell 2002; Cohn 2003).

CONCLUSIONS9

Pacific-Remittances.pmd 15/03/2005, 10:56 AM53

Remittances in the Pacific

54

These conclusions further suggest a number of policy recommendations,however, conclusions drawn mainly from dated studies in the relatively largestates of Samoa and Tonga cannot necessarily be generalized for other statesin the region. Indeed, the financial systems of Pacific island countries arequite different both in the extent of financial development and in institutionalstructure due to unique mixes of cultures and colonial and post colonialhistories. To some extent, however, the examples of Samoa and Tonga doprovide financial templates of remittance dependent economies that can beused as a good starting point for understanding the situation in other smallcountries in the region.

Several of these conclusions indicate that further in-depth research shouldbe undertaken. Few detailed studies of migration and remittances in the islandPacific are statistically significant, and many were done more than a decadeago and may no longer be valid. Two of the few studies that discussremittances in this century note that the wives of Kiribati seamen regularlycommunicated with their husbands by fax or telephone but that e-mail wasfast becoming their first preference (Clark 2004; Dennis 2003), opportunitiesthat were not open to migrants or recipients a decade ago. How new electronicforms of communication have influenced actual remittance flows has yet tobe explored in detail (cf. Morton Lee 2003, 2004a,b). There have been virtuallyno studies in Fiji Islands and Tuvalu though remittances are of considerableimportance in both.

Remarkably, not a single village study appears to have been conductedanywhere in the Pacific in the past decade that provides data on remittances.Firstly, follow-up studies in key villages where earlier studies were conductedto indicate potentially significant changes would be useful. Secondly, moredetailed studies would be useful in FSM, Niue, and the Cook Islands wherethere have been no recent studies, and in Kiribati and Tuvalu where therelationship between poverty, remittances, other income sources, and ruralsafety nets is of critical importance. Thirdly, detailed follow-up surveys inSamoa and Tonga that would indicate potentially significant changes overthe past decade would be valuable. Without more adequate data, it willcontinue to be difficult to be precise about trends, outcomes, and policies.

Critical uncertainties about some basic issues remain, including the intentof remitters and how that is translated (and even whether remitters are moreconservative than recipients), the extent of inequality created by the migration-remittance nexus, and the extent to which remittances contribute to crowdingout local economic activities. This is scarcely surprising since similaruncertainties occur in broader and better-documented contexts (Jones 1998).The present conclusions, while suggesting that remittance levels would besensitive to policies affecting relative real interest rates, do not suggest that

Pacific-Remittances.pmd 15/03/2005, 10:56 AM54

Conclusions

55

policy would stimulate more domestic investment from remittance flows orthat the migrants (or their families) would necessarily make the bestentrepreneurs despite the use of remittances for investment. This raises thebroader issue of the importance of the general investment climate. Migrantsare unlikely to risk their capital in an investment in the home economy ifsafer alternatives exist elsewhere.

In view of these sorts of problems, policies should be geared more towardsencouraging migrants to become more active in domestic capital markets assaver-rentiers. Governments must offer savers competitive real interest ratesin order to accumulate loan funds to invest either in larger domestic projectsor to hold as overseas assets denominated in foreign currency at the bestpossible rate of return. If and when investment opportunities arise in the homeeconomies, such offshore funds could also provide an important source ofventure capital.

This kind of intervention has not hitherto been considered in any detailin island states partly because it is technically difficult in open economies(and where migrants are often settlers) and because remitters have beenuninterested in (or unaware of) the few existing schemes. Governments arealso unwilling to tamper with substantial remittance inflows (the “goose layingthe golden egg” syndrome) or be seen to restrict freedom of movement,privacy, and individual decision making. Moreover there is a strong beliefthat supporting remitters or recipients would favor those who have alreadybenefited substantially from the system, hence equity would be poorly served.Finally Pacific island governments are relatively weak and ineffective indeveloping and implementing many policies and in other circumstances havebeen discouraged from intervening in the private sector.

It is increasingly evident however that maximizing the benefits ofinternational migration is crucial since it is highly valued throughout the regionfor social and economic reasons. As long as considerable economic challengesface island states, as their population growth rates remain above worldaverages, as development prospects are few, as the possibility of decliningaid becomes more apparent, and as expectations rise, the ability to migratewill be crucial.

There is no question that remittances have contributed to developmentin various contexts and senses. Most studies have suggested that many Pacificisland households use migration and remittances to increase their incomeseven fostering obligations and implicit contracts. Poirine has argued thatfamily members are consciously and repeatedly optimizing their economicstatus over time while the informal family credit market of remittances is anefficient means of achieving the, “…highest returns on human capitalinvestments” (1998). It is clear, however, that in many circumstances decision

Pacific-Remittances.pmd 15/03/2005, 10:56 AM55

Remittances in the Pacific

56

making is less rational and informed than such models suggest. Indeed therelationship between remittances and fluctuations in the labor markets in hostcountries is poorly understood, and there are almost no data that considerskilled and unskilled migrants as distinct categories. Moreover, there is arange of evidence that suggests that remittances are increasingly demanddriven and a response to requests. However, even with imperfect knowledge,households are consciously making decisions in favor of the quantity andquality of education of children that boost their chances for migration (Brownand Connell 2004). Migration and remittances thus stem from and contributeboth directly and indirectly to human capital formation.

Anecdotal evidence points to the growing individualism of overseasmigrants, especially to the increasing numbers of second generation migrantsborn overseas and the reduced likelihood that they will send remittances tohome countries especially if they take up host country citizenship. Thus oneTongan observed of migrants to Australia, “People who were born here orwent to school here send remittances. People born there? No way!” (James1997). Many Tongans in Melbourne have lost interest in continued financialsupport of their overseas kin as their sense of kavenga (obligations) hasdeclined over time (Morton Lee 2004). Similarly, skilled Tongan migrants inSydney are increasingly stating that they no longer remit (Fusitu’a 2000).Data on Samoan and Tongan nurses in Australia, on the other hand, indicatethat skilled migrants sustain remittances at high levels and over long timeperiods, but that data dates from the mid-1990s and circumstances may nowhave changed.

Even if direct remittances decline, migrants still contribute to collectivefundraising endeavors that play a part in nation building (Morton Lee 2004).The flow of food in both directions strongly suggests the affective andsymbolic role of exchange, how it is embedded in social structures as muchas economic transactions, and thus the potential for long term flows tocontinue. Indeed as remittances are a critical element in building andmaintaining social networks, they are likely to be sustained beyond whateconomic principles might suggest.

As first generation family ties decline, as families are reunited in hostcountries, and as there are no dependents in home countries, remittances forthat household will decrease, but other households will begin the migrationprocess anew. It is not therefore surprising that several Pacific states haveargued strongly and lobbied hard for special migration legislation in hostcountries to sustain migration flows. If migration becomes more selective interms of skills (or in any other way), gains to Pacific island states are likely todecline. It has long been argued that concessionary migration schemes are apractical form of aid to smaller Pacific island states (Connell 1984). These

Pacific-Remittances.pmd 15/03/2005, 10:56 AM56

Conclusions

57

arguments center around the view that since labor is the most successfulexport, concessionary migration policies would be even more beneficial. “Likeany other export, in the right policy context labour export and remittancescan serve as a stimulus to economic development and a change in comparativeadvantage” (Appleyard and Stahl 1995).

Two counter arguments to selective migration policies are thatdevelopment (and aid) should be centered in the island states, and that recipientcountries should have non-discriminatory migration policies (Cuthbertsonand Cole 1995) though both such arguments have been largely bypassed bytime. Indeed Australia has even been forced to contemplate the possibility ofdeveloping a migration-remittance economy for the impoverished state ofNauru (Hallett 2004). Moreover, these arguments take little account of laborneeds in host states. Short term labor migration has worked effectively inNew Zealand where workers from relatively poor parts of Fiji Islands havetaken up jobs, mainly in agriculture, that are unattractive to residents andhave returned to make substantial financial contributions to their homecommunities as a result (Levick and Bedford 1988). Workers who overstaytheir visas make similar contributions in Australia. Economic and institutionalchanges in present host countries do, however, place limits on this. Ultimatelythe future success of the migration-remittance nexus may lie in social,economic, and political changes in host states that open up labor markets.

As long as migration opportunities exist and offer prospects for higherearnings, it is perfectly rational for islanders to adopt strategies that maximizemigration prospects such as investing in children’s education. However,spending on education is treated as consumption (not investment) in nationalincome accounts, and in the intergenerational transfer model (Brown andPoirine 2004), remittances to parents are understood as returns on past (humancapital) investment. Once migration-induced investment in human capital istreated as a legitimate and rational use of resources, a number of the perceivedproblems with remittances and their use disappear, and the migration-remittances nexus is more obviously beneficial.

What is crucial for home countries is that long-term migrationopportunities continue. If improved education encourages children to stayand if retired emigrants are encouraged to return, for example, by ensuringaccess to land (a major challenge in most states) and by securing agreementswith host countries to preserve pension rights and any other retirement benefitsin their home countries, then remittances will be maximized (Poirine 1998).Moreover it is possible as James suggests for Tonga, that the export oftraditional wealth may continue to stimulate resource flows to Tonga (1997)even when remittances may dwindle.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM57

Remittances in the Pacific

58

Remittances in the Pacific are presently maintained for very long timeperiods, beyond what has hitherto been recorded in most other world regions,and in quite new socioeconomic contexts (Morton 1998, 1999). Indeed, themost striking conclusion of the most detailed studies is that remittances donot decline over time, emphasizing that migrants are ultimately motivated byfactors other than altruistic family support such as asset accumulation andinvestment at home as the intergenerational flow of remittances takes on amore individualistic element (Brown 1997, 1998). This may also reflect thepervasiveness of island social mores and some degree of discrimination inhost countries that increases the desire to maintain island social ties. Forwhatever combination of reasons, there is room for some degree of optimismthat remittance flows will not decline significantly in the near future, alongwith a degree of pessimism that flows will not continue indefinitely. Eventhat conclusion depends in large part on the continuity of migration flows. Itmight be expected that migration will continue when possible where publicsector employment is reduced, where wages and salaries remain low andunequal, where working conditions are sometimes difficult and hierarchical,where commodity prices have declined, and where there are many kinoverseas. Moreover, some skilled groups such as nurses are now recruitedinternationally.

As Muliaina has argued in the case of Samoa, if there is a continuedtightening of immigration policies by major Samoan hosts for whatever reason,“…the standard of living of rural Samoans, as opposed to urban dwellers,may be expected to decline in the next decade” (2001). If that is true ofSamoa, it is true of all other independent Pacific states where there is presentlya significant dependence on remittances. Moreover, Muliaina reached thatconclusion primarily for a Samoan village about 12 kilometers from the capitalwhere there were several business ventures and commuters to urbanemployment. Remote locations would face greater difficulties.

While international migration has had both positive and negative effectsin the Pacific, the significance of the positive effects (particularly increasedstandards of living) must be contrasted with the limited development potentialof many countries in the region and their failure to achieve significanteconomic growth or sustainable development. In the Pacific, as in theCaribbean, remittances have made a substantial contribution to householddevelopment but rather less to national development (Connell and Conway2000). In most of the South Pacific, the greater self-sufficiency that wouldfollow a decline of migration and remittances would be difficult and painful.As was argued for the small island of Rotuma, “With the prestige given to‘foreign’ goods, it is doubtful, therefore, that Rotumans would want to beself sufficient, even if it were a possibility” (Plant 1977). In Tikopia, Solomon

Pacific-Remittances.pmd 15/03/2005, 10:56 AM58

Conclusions

59

Islands, “…from such a level of dependence on imported goods it becomesdifficult to retreat without unease and a sense of deprivation” (Firth 1971). InPohnpei, too, villagers were not interested in adequate subsistence or even inthe right to subsistence, but rather desired, “…continued and increased accessto the goods and prestige provided by employment” (Petersen 1979). Morethan two decades later, these statements can be re-emphasized in a widercontext. Demand for migration and remittances is likely to be sustainedalongside rising expectations in conditions of limited national economicgrowth.

As preferential trade agreements disappear and barriers to internationalmigration become more selective and challenging, other forms of developmentand growth must be found and implemented, and remittances must be usedmore effectively. Many Pacific islanders, especially Polynesians, see migrationor overseas employment as a means of escaping limited domestic economicopportunities and maximizing their development options. In this context theneed to maximize the benefits from migration and remittances becomes evermore pressing.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM59

Ahlburg, D. 1991. Remittances and Their Impact: A Study of Tonga andWestern Samoa. Pacific Policy Papers No. 7. Canberra: NationalCentre for Development Studies, Australian National University.

———. 1995. Migration, Remittances and the Distribution of Income:evidence from the Pacific, Asian and Pacific Migration Journal, 4,157-168.

Ahlburg, D and Brown, R.P.C. 1997. Migrants’ Intentions to Return andCapital Transfers: a Study of Tongans and Samoans in Australia,Journal of Development Studies, 35, 125-151.

Ahlburg, D and Levin, M. 1990. The North East Passage: A Study of PacificIslander Migration to American Samoa and the United States,Canberra, National Centre for Development Studies.

Alexeyeff, K. 2004. Love Food: Exchange and Sustenance in the Cook IslandsDiaspora, Australian Journal of Anthropology, 15, 68-79.

Appleyard, R and Stahl, C. 1995. South Pacific Migration. New ZealandExperience and Implications for Australia, Canberra, AusAID.

Athukorala, P. 1993. Statistics on Asian Labour Migration: Review of Sources,Methods and Problems, in ILO ARTEP, Ed,. International LabourMigration Statistics and Information Networking in Asia, New Delhi,UNDP/ILO ARTEP.

Australia. 2003. A Pacific engaged – Australia’s relations with Papua NewGuinea and the island states of the south-west Pacific, Parliamentof Australia. Foreign Affairs Defence and Trade ReferencesCommitteee, Canberra.

Auty, R. 991. Sustaining Development in Mineral Economies: the ResourceCurse Thesis, London, Routledge.

Bedford, R. 1980. Demographic Processes in Small Islands: The Case ofInternal Migration, in H.C. Brookfield, Ed., Population-EnvironmentRelations in Tropical Islands: The Case of Eastern Fiji, Paris:UNESCO.

eferencesR

Pacific-Remittances.pmd 15/03/2005, 10:56 AM61

Remittances in the Pacific

62

Bellam, M.E.P. 1964. The Melanesian in Town, unpub. M.A. thesis, VictoriaUniversity of Wellington.Bertram, G. (1986) SustainableDevelopment in South Pacific Micro-economies, WorldDevelopment, 14, 809 22.

Bertram, G. 1999a. The MIRAB Model Twelve Years On, The ContemporaryPacific, 11, 105-138.

——.1999b Economy, in M. Rapaport, ed., The Pacific Islands: Environmentand Society, Honolulu, Bess Press, 337-352.

Bertram, G. and R.F. Watters. 1985. The MIRAB Economy in South PacificMicrostates, Pacific Viewpoint, 26, 497 520.

Besnier, N. 2004. Consumption and Cosmopolitanism: Practicing Modernityat the Second-Hand Marketplace in Nuku’alofa, Tonga,Anthropological Quarterly, 77, 7-45.

Borovnik, M. 2003. Seafarers in Kiribati—Consequences of InternationalLabour Circulation, unpub. PhD thesis, University of Canterbury,Christchurch.

———. 2004. Seafarer remittances to Kiribati: where do the benefits fall,unpub. paper to ‘Beyond MIRAB: the Political Economy of SmallIslands in the 21st Century’, Victoria University, Wellington.

Boyd, D. 1990. New Wealth and Old Power: Circulation, Remittances andthe Control of Inequality in an Eastern Highlands community, PapuaNew Guinea, in J. Connell, Ed., Migration and Development in theSouth Pacific. Ed NCDS Pacific Research Monograph No. 24.Canberra: National Centre for Development Studies, AustralianNational University.

Brown, R.P.C. 1994. Migrants’ Remittances, Savings and Investment in theSouth Pacific, International Labour Review, 133, 347 67.

———.1995. Hidden Foreign Exchange Flows: Estimating UnofficialRemittances to Tonga and Western Samoa, Asia Pacific MigrationJournal, 4, 35-54.

———.1997. Estimating Remittance Functions for Pacific Island Migrants,World Development, 25, 613-626.

———. 1998. Do migrants’ remittances decline over time? Evidence fromTongans and Western Samoans in Australia, The ContemporaryPacific, 10, 107-151.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM62

References

63

Brown, R.P.C. and Connell, J. 1993. The Global Flea Market: Migration,Remittances and the Informal Economy in Tonga, Development andChange, 24, 611 647.

———. 2004. The Migration of Doctors and Nurses from South Pacific IslandNations, Social Science and Medicine, 58, 2193-2210.

Brown, R.P.C. and Poirine, B. 2005. Intergenerational Transfers with Impurealtruism: an Analysis of Migrants’ Remittances, InternationalMigration Review, 39 (in press).

Browne, C. 1995. Pacific Island IMF Member Countries: Recent EconomicDevelopment and Medium-Term Prospects, Washington, IMF.

Campbell, I.C. 1992. A Historical Perspective on Aid and Dependency, PacificStudies, 15, 59 75.

Carrier, J. 1984. Education and Society in a Manus Village. EducationResearch Unit Report No.47, Port Moresby: University of PapuaNew Guinea.

Chambers, K. and Chambers, A. 2001. Unity of Heart, Culture and Changein a Polynesian Atoll Society, Waveland Press, Prospect Heights.

Chapman, M. and Prothero, R. 1985. Circulation in Population Movement.London: Routledge and Kegan Paul.

Clark, P. 2004. The economic impact of contracted labour upon the livelihoodsof small Pacific Island States: an examination of the expenditurepatterns of I-Kiribati and Tuvaluan seafarers and their dependents,unpub. Masters of Social Planning and Development thesis,University of Queensland.

Cohn, P. 2003. Economic Independence through Expansion of Private SectorEnterprise – the ‘Prescriptive Unreality’ of Niue’s DevelopmentPlanning, Melbourne University Private, School of DevelopmentStudies Working Paper No. 11, Melbourne.

Connell, J. 1980. Remittances and Rural Development: Migration,Dependency and Inequality in the South Pacific. National Centrefor Development Studies Occasional Paper No. 22, AustralianNational University, Canberra.

———.1983. Migration, Employment and Development in the South Pacific,Country Report No. 18. Tonga, Noumea, South Pacific Commission.

———.1984. Kiribati and Tuvalu: a suitable case for...?, Pacific Aid ResearchNewsletter, 5, September, 8-10.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM63

Remittances in the Pacific

64

———.1986. Population, Migration and Problems of Atoll Development inthe South Pacific, Pacific Studies, 9, 41–58.

———.1987. Migration, Rural Development and Policy Formation in theSouth Pacific, Journal of Rural Studies, 3, 105–121.

———.1988. Sovereignty and Survival: Island Microstates in the ThirdWorld. Research Monograph No.3. University of Sydney:Department of Geography.

——. Ed. 1990. Migration and Development in the South Pacific, PacificResearch Monograph No. 24. Canberra: National Centre forDevelopment Studies.

——. 1991. Migration and Remittances in the South Pacific Forum IslandCountries. Department of Geography, Sydney: University of Sydney.

——. 1999. Environmental Change, Economic Development and Emigrationin Tuvalu, Pacific Studies, 22, 1-20.

——. 2003a. Losing ground? Tuvalu, the greenhouse effect and the garbagecan, Asia Pacific Viewpoint, 44, 89-108.

——. 2003b. Island Dreaming: the Contemplation of Pacific Paradise, Journalof Historical Geography, 29, 554-582.

——. 2004. The Migration of Skilled Health Professionals: from the PacificIslands to the World, Asian and Pacific Migration Journal, 13 (2),155-177.

Connell, J. and Brown, R.P.C. 2004. The Remittances of Migrant Tonganand Samoan Nurses in Australia, Human Resources for Health, 2(2). (online)

Connell, J. and Conway, D. 2000. Migration and Remittances in IslandMicrostates: A Comparative Perspective on the South Pacific andthe Caribbean, International Journal of Urban and RegionalResearch, 24, 52-78.

Connell, J. and J. Lea. 1992. My Country Will Not Be There! Global Warming,Development and the Planning Response in Small Island States,Cities, 9, 295 309.

Connell, J and J Lea. 2002. Urbanisation in the Island Pacific. TowardSustainable Development, London, Routledge.

Cowling, W. 1990. Motivations for Contemporary Tongan Migration, in P.Herda, J. Terrell and N. Gunson, Eds. Tongan Culture and History,Canberra, Australian National University.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM64

References

65

Curson, P. 1979. Migration, remittances and social networks among CookIslanders, Pacific Viewpoint, 20, 185-198.

Cuthbertson, S and R. Cole. 1995. Population Growth in the South PacificIsland States, Canberra, Australian Government Publishing Service.

Dennis, J. 2003. Pacific Island Seafarers. A study of the economic and socialimplications of seafaring on dependants and communities,Secretariat of the Pacific Community, Suva.

Duncan, R. 1994. On Achieving Sound and Stable Economic Policies in thePacific Islands, Pacific Economic Bulletin, 9 (1), 21-25.

Evans, M. 1996. Is Tonga’s MIRAB economy sustainable? A View from theVillage, and a View without It. unpub. ms, University of NorthernBritish Columbia, Prince George.

Evans, M. 2001. Persistence of the Gift. Tongan Tradition in TransnationalContext, Wilfred Laurier University Press, Waterloo.

Faeamani, S. 1995. The Impact of Remittances on Rural Development inTongan Villages, Asian and Pacific Migration Journal, 4, 139-156.

Fairbairn, T.I.J. 1991a. The Western Samoan Economy: Prospects for Recoveryand Long term Growth. AIDAB, International Development IssuesNo.17.Canberra: Australian Government Publishing Service.

———.1991b. The Tongan Economy: Setting the Stage for AcceleratedGrowth. AIDAB, International Development Issues No.22.Canberra: Australian Government Publishing Service.

———.1993. Remittance Income. Its Importance for some Pacific IslandCountries and Implications for Production, in G. McCall and J.Connell, Eds., A World Perspective on Pacific Islander Migration,UNSW Centre for South Pacific Studies, Monograph No. 6, Sydney,311-318.

Faiva, O. 1989. The Tongans in Manly Warringah, Manly Hospital andCommunity Services. Sydney.

Finau, P. 1994. How Emigration Affects the Home Country, in G. McCalland J.Connell Eds, A World Perspective on Pacific IslanderMigration, Centre for South Pacific Studies Monograph No.6,University on New South Wales, Sydney.

Firth, R. 1971. Economic Aspects of Modernization in Tikopia, in L.R. Hiattand C. Jayawardena. Eds., Anthropology in Oceania, Sydney, Angusand Robertson.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM65

Remittances in the Pacific

66

Fisk, E.K. 1975. The response of nonmonetary production units to contactwith the exchange economy, in L. Reynolds Ed. Agriculture inDevelopment Theory, New Haven, Yale University Press, 53-83.

Forsyth, D.J.C. 1992. Migration and Remittances in the South Pacific ForumIsland Countries. Monograph, Department of Economics, Universityof the South Pacific, Suva.

Foster, J. 1995. The relationship between Remittances and Savings in SmallPacific Island states: Some Econometric Evidence, Asian and PacificMigration Journal, 4, 117-138

Fusitu’a, P. 2000. personal comment, 11 November.

Fuka, M.L.A. 1985. The Auckland Tongan Community and OverseasRemittances, unpub. M.A. thesis, University of Auckland.

Gailey, C.W. 1992a. A Good Man is Hard to Find, Critique of Anthropology,12, 47 74.

——. 1992b State Formation, Development and Social Change in Tonga, inSocial Change in the Pacific Islands, in A. Robillard, E., London:Kegan Paul International.

Gordon, J and P. Gupta. 2004. Non-Resident Deposits in India: In Search ofReturn?, Economic and Political Weekly, 11 September, 4165-4174.

Grieco, E. 2003. The Remittance Behavior of Immigrant Households:Micronesians in Hawaii and Guam, LFB Scholarly Publishing LLC,New York.

Grieco, E. 2004. Will Migrant Remittances Continue Through Time? A NewAnswer to an Old Question, International Journal of MulticulturalStudies (in Press)

Haberkorn, G. 1989. Port Vila: Transit Station or Final Stop? Pacific ResearchMonograph No.21. Canberra, National Centre for DevelopmentStudies, Australian National University.

Halatuituia, S. 2001. Tonga’s Contemporary Land Tenure System: Realityand Rhetoric, unpub. Ph.D thesis, University of Sydney, Sydney.

Hallett, D. 2004. personal comment, 20 June.

Hardaker, J., et al. 1987. Smallholder Agriculture in Tonga. Armidale:University of New England.

Hau’ofa, E. 1994. Our sea of islands, The Contemporary Pacific, 6, 147-161.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM66

References

67

Hayes, G. 1993. ‘Mirab’ processes and Development on Small Pacific Islands:A Case Study from the southern Massim, Papua New Guinea, PacificViewpoint, 34, 153 178.

Helweg, W.H. 1983. Emigrant Remittances: Their Nature and Impact on aPunjab Community, New Community, 10, 435 43.

Hess, J. 2004. Wave and Reflection: Charting Marshallese participation inGlobalizing Processes, in V. Lockwood, Ed, Globalization andCulture Change in the Pacific Islands, Pearson, New Jersey,182-199.

Heyn, J. A. 2003. Migration and Development on Niue Island, unpub. M.Sc.thesis, University of Montana.

Hooker, K. and J. Varcoe, 1999. Migration and the Cook Islands, in J. Overtonand R. Scheyvens, Eds, Strategies for Sustainable Development.Experiences from the Pacific, Sydney, UNSW Press, 91-99.

Hooper, A. 1993. The MIRAB Transition in Fakaofo, Pacific Viewpoint, 34,241-264.

James, K.E. 1991. Migration and Remittances: A Tongan Village Perspective,Pacific Viewpoint, 32, 1-23.

James, K. 1993a. The Rhetoric and Reality of Change and Development inSmall Pacific Communities, Pacific Viewpoint, 34, 135-152.

———.1993b. Cash and Kin. Aspects of Migration and Remittance fromthe Perspective of a Fishing Village in Vava’u, Tonga, in G. McCalland J. Connell Eds., A World Perspective on Pacific IslanderMigration, UNSW Centre for South Pacific Studies Monograph No.6, Sydney, 359-374.

———.1997. Reading the Leaves: The Role of Tongan Women’s TraditionalWealth and other ‘Contraflows’ in the Processes of ModernMigration and Remittance, Pacific Studies 20, 1-27.

Jones, R. 1998. The Renewed Role of Remittances in the New World Order,Economic Geography ,74, 1-7.

Karakita, Y. 1997. Prior to MIRAB?: Remittances and Inter-Island Relationsin Woleai Atoll, Yap State, Federated States of Micronesia, in K.Sudo and S. Yoshida, Eds, Contemporary Migration in Oceania:Diaspora and Network, Japan Centre for Area Studies, Osaka, 11-24.

Kardar, S. 1992 Exchange and Payment Reforms in Pakistan. New Delhi:UNDP/ILO ARTEP.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM67

Remittances in the Pacific

68

Kazi, S. 1989. Domestic Impact of Overseas Migration: Pakistan, in R. Amjad,Ed., To the Gulf and Back: Studies on the Economic Impact of AsianLabour Migration, New Delhi: ILO/ UNDP.

La Plagne, P., M. Treadgold, and J. Baldry. 2001. A Model of Aid Impact insome South Pacific Microstates, World Development, 29, 365-383.

Latailakepa, S. 1997. Female Migration in Tonga: Motives and Consequences,unpub. MA thesis, James Cook University, Townsville.

Levick, W and R. Bedford. 1988. Fiji labour migration to New Zealand inthe 1980s, New Zealand Geographer, 44, 14-21.

Liki, A (1994) E Tele A’a o le Tagata: Career Choices of Samoan Professionalswithin and beyond their Nu’u Moni, unpub. M.A. thesis, Universityof the South Pacific, Suva.

Liki, A. 2001. Moving and rootedness: the paradox of the brain drain amongSamoan professionals, Asia-Pacific Population Journal, 16 (1),67-84.

Loomis, T. 1990. Cook Island Remittances: Volumes, Determinants and Uses,in J. Connell, Ed., Migration and Development in the South Pacific.Pacific Research Monograph No. 24. Canberra: National Centrefor Development Studies, Australian National University.

MacMaster, J. 1993 Strategies to Stimulate Private Sector Development inthe Pacific Island Economies, in R.Cole and S. Tambunlertchai,Eds , The Future of Asia-Pacific Economies, Canberra, NationalCentre for Development Studies, ANU.

Macpherson, C. 1992. Economic and Political Restructuring and theSustainability of Migrant Remittances: The Case of Western Samoa,The Contemporary Pacific, 4, 109 136.

———.1994. Changing Patterns of Commitment to Island Homelands: acase study of Western Samoa, Pacific Viewpoint, 17, 83-116.

———.2004. Transnationalism and Transformation in Samoan Society, inV. Lockwood, Ed, Globalization and Culture Change in the PacificIslands, Pearson, New Jersey, 165-181.

Mahmud, W. 1989 The Impact of Overseas Labour Migration on theBangladesh Economy, in R. Amjad, Ed., To the Gulf and Back:Studies on the Economic Impact of Asian Labour Migration, NewDelhi: ILO/UNDP.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM68

References

69

Marcus, G.E. 1993 Tonga’s Contemporary Globalizing Strategies: Tradingon Sovereignty Amidst International Migration.” in ContemporaryPacific Societies, Eds. T. Harding and B. Wallace. Englewood Cliffs:Prentice Hall.

———.1981. Power on the Extreme Periphery: The Perspective of TonganElites in the Modern World System, Pacific Viewpoint, 22, 48-64.

Marshall, M. 2004. Namoluk Beyond The Reef. The Transformation of aMicronesian Society, Westview, Boulder.

Mataio, M. 1991 Scope for Sustainable Development in the Context of theMIRAB Hypothesis: the Case of the Cook Islands, unpub. M.Econ.thesis, University of New England, Armidale.

Mecartney, S. 2000. Blacksands Settlement: Towards Urban Permanence inVanuatu, unpub. M.A. thesis, University of Sydney.

Meleisea, M. 1992. Change and Adaptations in Western Samoa, MacmillanBrown Centre for Pacific Studies, University of CanterburyChristchurch,

Miles, R., et al. 1992. Employment in Tonga: Present and Potential. Suva,Fiji: ILO/UNDP/AIDAB.

Morauta, L. 1985. Urban Movement and Rural Identity: A Papua New GuineaExample. in M. Chapman, Ed,. Mobility and Identity in the IslandPacific, Pacific Viewpoint (special issue), 26, 221 241.

Morton, H. 1998. Creating Their Own Culture: Diasporic Tongans, TheContemporary Pacific 10, 1-30.

Morton, H. 1999. Islanders in Space: Tongans Online, in R.King and J.ConnellEds, Small Worlds, Global Lives. Islands and Migration, Pinter,London, 235-253.

Morton Lee, H. 2003. Tongans Overseas. Between Two Shores, Honolulu,University of Hawaii Press.

———2004a. All Tongans Are Connected: Tongan Transnationalism, in V.Lockwood, Ed, Globalization and Culture Change in the PacificIslands, New Jersey, Pearson, 133-148.

———. 2004b. ‘Second generation’ Tongan transnationalism: hope for thefuture, unpub. paper to ‘Beyond MIRAB: The Political Economyof Small Islands in the 21st Century’, Victoria University, Wellington.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM69

Remittances in the Pacific

70

Muliaina, T. 2001. Remittances, the Social System and Development inSamoa, in V. Naidu, E. Vasta and C. Hawksley, Eds., Current Trendsin South Pacific Migration, Asia Pacific Migration ResearchNetwork Working Paper No. 7, Centre for Asia Pacific SocialTransformation Studies, Wollongong, 20-40.

Munro, D. 1990. Transnational corporations of kin and the MIRAB system:the case of Tuvalu, Pacific Viewpoint, 31, 63-66.

Ogden, M. 1994. MIRAB and the Republic of the Marshall Islands, Isla, 2,237-272.

O’Meara, T. 1990. Samoan Planters. Tradition and Economic Developmentin Polynesia. Fort Worth: Holt, Rinehart and Winston.

Overton,J, Murray, W and I. Ali. 1999. Commodity Production andUnsustainable Agriculture, in J.Overton and R.Scheyvens, Eds,Strategies for Sustainable Development. Experiences from thePacific, Sydney, UNSW Press, 168-181.

Petersen, G. 1979. External Politics, Internal Economics and Ponapean SocialFormation, American Ethnologist, 6:25-40.

Pirie, P. (1960) The Population of Western Samoa. A Preliminary ReportBased on the 1956 Census, Canberra.

Pirie, P. and W. Barrett. 1962. Western Samoa. Population, Production andWealth, Pacific Viewpoint, 3:63 96.

Plant, C. 1977. The Development Dilemma, in Rotuma: Split Island. Ed.C.Plant. Suva: Institute of Pacific Studies.

Poirine, B. 1997. A Theory of Remittances as an Informal Family LoanAgreement, World Development, 25, 221-241.

———.1998. Should we hate or Love MIRAB?, The Contemporary Pacific,10, 1998, 65-105.

Pollard, S. 1995 Pacific Economic Policy: To Invest or to Protect?, EastWest Center Pacific Islands Development Program Working PaperNo.11, Honolulu.

Prasad, N. 2003. Small Islands’ Quest for Economic Development, Asia-Pacific Development Journal, 10 (1), June, 47-67.

Quibria, M.G. 1986. Migrant Workers and Remittances: Issues for AsianDeveloping Countries, Asian Development Review, 4, 78 99.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM70

References

71

Quibria, M.G. and M. Thant. 1988. International Labor Migration, Emigrants’Remittances, and Asian Developing Countries: Economic Analysisand Policy Issues, in M.Dulta, Ed, Research in Asian EconomicStudies, Greenwich, Connecticut: Jal Press.

Rensel, J. 1993. The Fiji Connection: Migrant Involvement in the Economyof Rotuma, Pacific Viewpoint, 34, 215-240.

Russell, S.S. 1986. Remittances from International Migration: a Review inPerspective, World Development 14, 677-696.

———.1992 Migrant remittances and development, International Migration,30, 267-287.

Saith, A. 1989. Macroeconomic Issues in International Labour Migration: AReview, in R. Amjad, Ed., To The Gulf and Back, New Delhi, ILO/UNDP.

Scott, G. G. 2003. Situating Fijian Transmigrants: Towards RacialisedTransnational Social Spaces of the Undocumented, InternationalJournal of Population Geography, 9, 181-198.

Scott, J. 1976. The Moral Economy of the Peasant, New Haven, YaleUniversity Press.

Sevele, F. 1973. The Importance of Agriculture in Pacific Island Development.Paper presented at the Episcopal Conference, Wellington.

Shankman, P. 1976. Migration and Underdevelopment: The Case of WesternSamoa, Boulder, Westview.

Simati, A. and J. Gibson. 2001. Do remittances decay? Evidence fromTuvaluan migrants in New Zealand, Pacific Economic Bulletin, 16(1), 55-63.

Small, C. 1995. The Birth and Growth of a Polynesian Women’s ExchangeNetwork, Oceania, 65, 234-256.

Small, C (1997) Voyages: From Tongan Villages to American Suburbs, Ithaca,Cornell University Press.

So’o, A. 1998. The Price of Election campaigning in Samoa, in P. Larmour,Ed, Governance and Reform in the South Pacific, National Centrefor Development Studies, Australian National University, Canberra,289-304.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM71

Remittances in the Pacific

72

Stanwix, C. and John Connell. 1995. To the Islands: the Remittances of Fijiansin Sydney, Asian and Pacific Migration Journal, 4, 69-88.

Straubhaar, T. 1986 The Determinants of Workers’ Remittances: The Caseof Turkey, Weltwirtschaftliches Archiv, 122, 728 739.

Sturton, M. 1992. Tonga: Development Through Agricultural Exports, PacificIslands Development Program Economic Report No.4, Honolulu.

Sudo, K. 1997. Expanding International Migration by the Tongan People:Strategies and Socio-cultural Effects on the Homeland, in K. Sudoand S. Yoshida, Eds, Contemporary Migration in Oceania: Diasporaand Network, Japan Centre for Area Studies, Osaka, 101-112.

Swamy, G. 1981. International Migrant Workers’ Remittances: Issues andProspects. World Bank Staff Working Papers No.481. Washington,DC: World Bank.

Tingsabadh, C. 1989. Maximising Development Benefits from LabourMigration: Thailand, in R. Amjad, Ed., To the Gulf and Back: Studieson the Economic Impact of Asian Labour Migration, New Delhi:ILO/UNDP.

Tisdell, C. 2002 Globalisation, development and poverty in the Pacific Islands.The situation of the least developed nations, International Journalof Social Economics, 29, 909-922.

Tonga, Kingdom of. 1991. Sixth Development Plan 1991-1995, Nuku’alofa,Central Planning Department.

Tongamoa, T. 1987 Migration, Remittances and Development: A TonganPerspective, unpub M.A. thesis, University of Sydney.

———.1990. International Migration, Remittances and Tonga, Review, 18,12 18.

Underhill, Y. 1989. Population mobility as a household strategy: the case ofManihiki atoll, Cook Islands, unpub. MA thesis, University ofHawaii, Honolulu.

Van der Grijp, P. 1997 Leaders in Squash Export: Entrepreneurship and theIntroduction of a new cash crop in Tonga, Pacific Studies, 20, 29-62.

———.1999. Modernisation, mondialisation et identite Polynesienne. Lemodele ‘MIRAB’ dans le Pacifique, La Pensee, 318, 83-95.

Walker, A. and R. P. C Brown. 1995. From Consumption to Savings?Interpreting Tongan and Western Samoan Sample Survey Data onRemittances, Asian and Pacific Migration Journal, 4, 89-116.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM72

References

73

Western Samoa. 1975. Western Samoa’s Third Five Year Development Plan,1975 1979. Apia.

Winthorpe, M. 2004. Words from Whitesands. Coral Reef Management inTanna, Vanuatu, unpub. B.Sc thesis, University of Sydney, Sydney.

World Bank. 1981. Labor Migration from Bangladesh to the Middle East.World Bank Staff Working Papers No.454. Washington, DC: TheWorld Bank.

———.1990. Towards Higher Growth in Pacific Island Economies,Washington.

———.1993. Pacific Island Economies: Toward Efficient and SustainableGrowth,Vols.1,6 and 8.Washington, DC: The World Bank.

———.1991. Pacific Island Economies: Towards Higher Growth in the l990s.Washington, DC: The World Bank.

Wurtzburg, S. 2004. Creating a New Zealand-Styled Fa’a-Samoa: SamoanIdentity in Christchurch, New Zealand Sociology, 19, 50-76.

Pacific-Remittances.pmd 15/03/2005, 10:56 AM73


Recommended